Chiebot-Mirror
/
opencv
mirror of https://github.com/opencv/opencv.git
8
0
Fork 0
Code Issues Projects Releases Wiki Activity
Open Source Computer Vision Library https://opencv.org/
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
13810 Commits
7 Branches
126 Tags
3.0 GiB
Tag: Branch: Tree: 964b260937
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '964b260937'
${ noResults }
opencv/samples/cpp/trained_classifierNM1.xml

4047 lines
134 KiB
Raw Normal View History Unescape

Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<?xml version="1.0"?>
<opencv_storage>
<boost type_id="opencv-ml-boost-tree">
<boosting_type>RealAdaboost</boosting_type>
<splitting_criteria>Gini</splitting_criteria>
<ntrees>100</ntrees>
<weight_trimming_rate>1.</weight_trimming_rate>
<is_classifier>1</is_classifier>
<var_all>5</var_all>
<var_count>4</var_count>
<ord_var_count>4</ord_var_count>
<cat_var_count>0</cat_var_count>
<training_params>
<use_surrogates>0</use_surrogates>
<max_categories>10</max_categories>
<max_depth>1</max_depth>
<min_sample_count>10</min_sample_count>
<cross_validation_folds>0</cross_validation_folds>
<priors type_id="opencv-matrix">
<rows>1</rows>
<cols>2</cols>
<dt>d</dt>
<data>
1. 1.</data></priors></training_params>
<var_idx type_id="opencv-matrix">
<rows>1</rows>
<cols>4</cols>
<dt>i</dt>
<data>
1 2 3 4</data></var_idx>
<var_type>
0 0 0 0</var_type>
<cat_count type_id="opencv-matrix">
<rows>1</rows>
<cols>1</cols>
<dt>i</dt>
<data>
2</data></cat_count>
<cat_map type_id="opencv-matrix">
<rows>1</rows>
<cols>2</cols>
<dt>i</dt>
<data>
1 2</data></cat_map>
<trees>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-8.7333809708193622e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>8.9537560939788818e-01</quality>
<le>8.5429996252059937e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>281</sample_count>
<value>6.7336438045685210e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1578</sample_count>
<value>-1.6797288877447023e+00</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>6.8833827526759232e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>6.0157060623168945e-01</quality>
<le>1.5000000000000000e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1728</sample_count>
<value>-2.3830233314576746e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>131</sample_count>
<value>1.3270520927417431e+00</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.1546319456101495e-14</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>6.0729509592056274e-01</quality>
<le>1.0963299870491028e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>781</sample_count>
<value>2.8488485626451149e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1078</sample_count>
<value>-1.0290340538001230e+00</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>8.8817841970011734e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.8338499069213867e-01</quality>
<le>9.5209401845932007e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1331</sample_count>
<value>-4.6437394337436783e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>528</sample_count>
<value>4.0542316486679569e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.5931700403371249e-14</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>3</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.5465269088745117e-01</quality>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<le>3.</le></_></splits></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>856</sample_count>
<value>5.0630658667623052e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1003</sample_count>
<value>-2.3844591468675169e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-4.0523140398818379e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.5272746086120605e-01</quality>
<le>5.0000000000000000e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1046</sample_count>
<value>-3.5086409511139033e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>813</sample_count>
<value>3.2363636038761090e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.5527136788004883e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.4958724975585938e-01</quality>
<le>1.6267855167388916e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1811</sample_count>
<value>-1.0723654593445622e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>48</sample_count>
<value>1.6463996332936000e+00</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.5543122344752215e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.5254465341567993e-01</quality>
<le>7.4758999049663544e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>149</sample_count>
<value>1.0273978441887677e+00</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1710</sample_count>
<value>-1.3682146780021626e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.1990408665951547e-14</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.4537707567214966e-01</quality>
<le>6.6345453262329102e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>455</sample_count>
<value>-1.0072489119231773e+00</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1404</sample_count>
<value>1.1925375544044788e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-7.1054273576010523e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.2146106958389282e-01</quality>
<le>2.5000000000000000e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1791</sample_count>
<value>-4.7305821386203636e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>68</sample_count>
<value>1.5160735804641536e+00</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.7747582837255180e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>3</var>
<quality>5.2081108093261719e-01</quality>
<le>5.</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1690</sample_count>
<value>9.1796156088740075e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>169</sample_count>
<value>-4.9059916762377265e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-7.1054273576010523e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.2672868967056274e-01</quality>
<le>1.0963299870491028e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>781</sample_count>
<value>1.0385059515501381e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1078</sample_count>
<value>-5.7169325870124055e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.1102230246251577e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>3</var>
<quality>5.2609002590179443e-01</quality>
<le>3.</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>856</sample_count>
<value>3.2591023278276837e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1003</sample_count>
<value>-1.6727738809775036e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.5527136788004883e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.1409310102462769e-01</quality>
<le>7.1857005357742310e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>114</sample_count>
<value>6.4622038086485267e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1745</sample_count>
<value>-4.9566471799398701e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-7.4940054162198634e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.1351374387741089e-01</quality>
<le>1.4143149554729462e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1571</sample_count>
<value>-4.6424296372866977e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>288</sample_count>
<value>6.6639156578772774e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.6645352591003686e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.2261954545974731e-01</quality>
<le>1.0177600383758545e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>590</sample_count>
<value>1.3091283509473614e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1269</sample_count>
<value>-3.6264100845476954e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>0.</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.2181720733642578e-01</quality>
<le>8.9620999991893768e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>352</sample_count>
<value>-2.4183425262980360e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1507</sample_count>
<value>1.8605210755965146e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.1102230246251554e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0836616754531860e-01</quality>
<le>1.0963299870491028e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>781</sample_count>
<value>5.1954271677409611e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1078</sample_count>
<value>-3.3426585985028628e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.7747582837255180e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0880891084671021e-01</quality>
<le>1.1348484754562378e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1664</sample_count>
<value>-7.4858358240636261e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>195</sample_count>
<value>2.4031047447238377e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.3322676295501896e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.1081919670104980e-01</quality>
<le>1.2661495208740234e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1730</sample_count>
<value>5.6719022662652843e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>129</sample_count>
<value>-4.0229379476194682e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>5.5511151231257519e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0904816389083862e-01</quality>
<le>2.3689103126525879e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1838</sample_count>
<value>-1.9140961693183144e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>21</sample_count>
<value>1.7878855228391193e+00</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-9.2703622556201423e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0664407014846802e-01</quality>
<le>2.2119045257568359e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1836</sample_count>
<value>1.8191803445960011e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>23</sample_count>
<value>-9.2985231739132745e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.1102230246251577e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0936371088027954e-01</quality>
<le>1.2460045814514160e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1724</sample_count>
<value>-5.7918483977216939e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>135</sample_count>
<value>3.3577728337370422e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.4424906541753385e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0752711296081543e-01</quality>
<le>9.5505046844482422e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1341</sample_count>
<value>1.1992717133073164e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>518</sample_count>
<value>-1.2680428752841727e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-4.8849813083507124e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0749230384826660e-01</quality>
<le>7.8355848789215088e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>764</sample_count>
<value>-2.4384603502165489e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1095</sample_count>
<value>6.2746713049518996e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.9984014443252778e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0703763961791992e-01</quality>
<le>6.3535496592521667e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>59</sample_count>
<value>5.7564577324623896e+00</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1800</sample_count>
<value>-1.4076159604991374e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.4512436177016777e-05</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0712978839874268e-01</quality>
<le>6.4906999468803406e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>61</sample_count>
<value>-2.8822016897573164e+00</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1798</sample_count>
<value>1.4394134992880541e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.4408920985006242e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.1009607315063477e-01</quality>
<le>6.9339498877525330e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>87</sample_count>
<value>1.6547355644710868e+00</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1772</sample_count>
<value>-2.1726950528099288e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.1102230246251577e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>3</var>
<quality>5.0710922479629517e-01</quality>
<le>7.</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1823</sample_count>
<value>1.7568593940350565e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>36</sample_count>
<value>-1.1252698341588880e+00</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.7763568394002536e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0767201185226440e-01</quality>
<le>8.0150999128818512e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>206</sample_count>
<value>2.7041738749866351e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1653</sample_count>
<value>-5.8184179168035623e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.3306690738754586e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0510478019714355e-01</quality>
<le>6.2720644474029541e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>374</sample_count>
<value>-4.5462988716181019e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1485</sample_count>
<value>2.3987712659202257e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-4.2188474935756130e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.0439256429672241e-01</quality>
<le>2.5000000000000000e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1791</sample_count>
<value>-1.5635519241144060e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>68</sample_count>
<value>6.3562787613239691e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>6.6613381477508951e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>3</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0438237190246582e-01</quality>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<le>3.</le></_></splits></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>856</sample_count>
<value>1.3081676957882793e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1003</sample_count>
<value>-6.7484142743913345e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.1102230246251554e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0724387168884277e-01</quality>
<le>1.1533749848604202e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>921</sample_count>
<value>4.0250806766256482e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>938</sample_count>
<value>-3.7703668263161400e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-5.8286708792821058e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0915646553039551e-01</quality>
<le>1.4143149554729462e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1571</sample_count>
<value>-3.3928131204983447e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>288</sample_count>
<value>6.0410410576585261e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.2188474935755767e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0659310817718506e-01</quality>
<le>1.4145749807357788e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1572</sample_count>
<value>2.0368450770510211e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>287</sample_count>
<value>-7.7093695859120004e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.3877787807814476e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0616967678070068e-01</quality>
<le>1.6267855167388916e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1811</sample_count>
<value>-2.4173163361906828e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>48</sample_count>
<value>5.6348494257358028e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-3.8857805861880494e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0677150487899780e-01</quality>
<le>1.2661495208740234e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1730</sample_count>
<value>4.0402378477571901e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>129</sample_count>
<value>-3.4888339817180891e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.4408920985006242e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0588327646255493e-01</quality>
<le>1.0797724723815918e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1608</sample_count>
<value>-6.8287991647037555e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>251</sample_count>
<value>1.7431967242459881e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.6629367034256354e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0476020574569702e-01</quality>
<le>9.2052650451660156e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1253</sample_count>
<value>1.0513502662517235e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>606</sample_count>
<value>-9.1138849679050823e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0690639019012451e-01</quality>
<le>9.0829002857208252e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1219</sample_count>
<value>-1.4636537853223561e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>640</sample_count>
<value>9.5332675553986543e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.2188474935755767e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0534439086914062e-01</quality>
<le>9.2052650451660156e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1253</sample_count>
<value>1.1293790129864213e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>606</sample_count>
<value>-9.5332675553986113e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.1649348980190600e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0435024499893188e-01</quality>
<le>9.3664005398750305e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>411</sample_count>
<value>8.4555682036086433e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1448</sample_count>
<value>-1.0351006034230040e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.2204460492503126e-16</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0704753398895264e-01</quality>
<le>9.9698998034000397e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>544</sample_count>
<value>-8.5042709024257673e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1315</sample_count>
<value>1.6769533771522532e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0574851036071777e-01</quality>
<le>1.0386450588703156e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>639</sample_count>
<value>5.3643508899729207e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1220</sample_count>
<value>-2.1791341988666471e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>0.</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0597953796386719e-01</quality>
<le>1.4143149554729462e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1571</sample_count>
<value>-2.3199011211312900e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>288</sample_count>
<value>5.7032019870719552e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0620901584625244e-01</quality>
<le>1.1533749848604202e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>921</sample_count>
<value>3.0260573388611104e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>938</sample_count>
<value>-4.3620446423872927e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0698792934417725e-01</quality>
<le>1.0789400339126587e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>742</sample_count>
<value>-4.7900146843168623e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1117</sample_count>
<value>3.0074316296999687e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0459891557693481e-01</quality>
<le>9.0829002857208252e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1219</sample_count>
<value>-1.1779072393798255e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>640</sample_count>
<value>7.8608870331688729e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.3322676295501861e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0399231910705566e-01</quality>
<le>3.3933651447296143e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>59</sample_count>
<value>1.3770357317310018e+00</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1800</sample_count>
<value>-9.0707243801608061e-03</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.6090241078691246e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0383007526397705e-01</quality>
<le>1.0386450588703156e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>639</sample_count>
<value>4.0916632352681503e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1220</sample_count>
<value>-1.8955755646648090e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0380486249923706e-01</quality>
<le>9.9698998034000397e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>544</sample_count>
<value>-5.8888670701558621e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1315</sample_count>
<value>1.3009980239826119e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.2204460492503083e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0560057163238525e-01</quality>
<le>9.4531498849391937e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>431</sample_count>
<value>8.6647133637941770e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1428</sample_count>
<value>-1.3032934253469194e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.3877787807814476e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0492835044860840e-01</quality>
<le>8.9620999991893768e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>352</sample_count>
<value>-1.0506312397672390e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1507</sample_count>
<value>9.4441739612300818e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-9.4368957093138405e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0368678569793701e-01</quality>
<le>9.2722505331039429e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>392</sample_count>
<value>7.6022467408684444e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1467</sample_count>
<value>-9.7486085972466607e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.1086244689504288e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0421810150146484e-01</quality>
<le>9.9444493651390076e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>536</sample_count>
<value>-6.3752797298286562e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1323</sample_count>
<value>1.3328967944519532e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.4408920985006242e-16</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0323176383972168e-01</quality>
<le>1.0177600383758545e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>590</sample_count>
<value>4.3831310206729070e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1269</sample_count>
<value>-1.4864401441489694e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-8.8817841970012602e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0349098443984985e-01</quality>
<le>1.0789400339126587e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>742</sample_count>
<value>-3.4075730824220232e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1117</sample_count>
<value>2.0791874238488089e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-4.4408920985006281e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0392138957977295e-01</quality>
<le>1.0963299870491028e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>781</sample_count>
<value>2.6986027157566792e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1078</sample_count>
<value>-2.9932407189155552e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0362765789031982e-01</quality>
<le>2.3689103126525879e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1838</sample_count>
<value>-9.2384036977927863e-03</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>21</sample_count>
<value>1.0592165467278325e+00</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.9984014443252857e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0543153285980225e-01</quality>
<le>2.2119045257568359e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1836</sample_count>
<value>1.5914551618823981e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>23</sample_count>
<value>-8.3405482949082688e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.5527136788004883e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0523018836975098e-01</quality>
<le>1.6267855167388916e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1811</sample_count>
<value>-1.9824411253917307e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>48</sample_count>
<value>5.8698394468331438e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-4.2188474935756130e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0369459390640259e-01</quality>
<le>1.2661495208740234e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1730</sample_count>
<value>2.8318382799977653e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>129</sample_count>
<value>-2.6718445379285333e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.2204460492503126e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0479084253311157e-01</quality>
<le>1.2460045814514160e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1724</sample_count>
<value>-3.7451522137317672e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>135</sample_count>
<value>2.6178246916992864e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>6.6613381477509353e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0358188152313232e-01</quality>
<le>1.2661495208740234e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1730</sample_count>
<value>2.7995062956094632e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>129</sample_count>
<value>-2.6178246916992898e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-5.5511151231257827e-17</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.0313872098922729e-01</quality>
<le>5.0000000000000000e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1046</sample_count>
<value>-8.7344555737785268e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>813</sample_count>
<value>7.2176940461025546e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.9984014443252857e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0293076038360596e-01</quality>
<le>1.2460045814514160e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1724</sample_count>
<value>-2.8268637265518849e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>135</sample_count>
<value>2.1045949672703843e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>4.2188474935755767e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>3</var>
<quality>5.0332474708557129e-01</quality>
<le>3.</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>856</sample_count>
<value>1.2230679161917310e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1003</sample_count>
<value>-5.4692581535459049e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.9984014443252857e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0446850061416626e-01</quality>
<le>1.1533749848604202e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>921</sample_count>
<value>2.3320665865473968e-02</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>938</sample_count>
<value>-4.0391049572091403e-01</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0324887037277222e-01</quality>
<le>8.0150999128818512e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>206</sample_count>
<value>1.6178738318585617e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1653</sample_count>
<value>-4.0534413064780048e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0377529859542847e-01</quality>
<le>1.4143149554729462e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1571</sample_count>
<value>-1.6597530926694307e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>288</sample_count>
<value>4.9094014377722600e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-3.1086244689504478e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0313144922256470e-01</quality>
<le>4.5337548851966858e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>110</sample_count>
<value>-8.1199669103862704e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1749</sample_count>
<value>9.3381880402274662e-03</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.7747582837255180e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0475740432739258e-01</quality>
<le>4.5797997713088989e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>113</sample_count>
<value>6.8212891092482997e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1746</sample_count>
<value>-1.6049266927207175e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0308030843734741e-01</quality>
<le>6.4906999468803406e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>61</sample_count>
<value>-1.8287112564925503e+00</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1798</sample_count>
<value>6.4869817401182221e-03</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>6.6613381477509353e-16</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0292497873306274e-01</quality>
<le>1.4145749807357788e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1572</sample_count>
<value>1.1075480383243890e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>287</sample_count>
<value>-5.8766713493584966e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0272303819656372e-01</quality>
<le>1.0132449865341187e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>577</sample_count>
<value>-4.4462051722396978e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1282</sample_count>
<value>1.2318755109692310e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0431984663009644e-01</quality>
<le>9.4531498849391937e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>431</sample_count>
<value>7.5132879390893734e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1428</sample_count>
<value>-1.1572265040778912e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0449663400650024e-01</quality>
<le>8.3246499300003052e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>254</sample_count>
<value>-1.3886315377806810e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1605</sample_count>
<value>6.5272198323505989e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.2204460492503126e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>1</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0269138813018799e-01</quality>
<le>8.0150999128818512e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>206</sample_count>
<value>1.4709439757444326e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1653</sample_count>
<value>-3.6874150180988774e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-6.6613381477509432e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.0234586000442505e-01</quality>
<le>5.5000000000000000e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1819</sample_count>
<value>-6.0288987564491008e-03</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>40</sample_count>
<value>1.0408276196291879e+00</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.2204460492503136e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.0341212749481201e-01</quality>
<le>4.5000000000000000e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1810</sample_count>
<value>1.0137680632639267e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>49</sample_count>
<value>-8.1653792840371642e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.2767564783189316e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0258362293243408e-01</quality>
<le>9.2052650451660156e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1253</sample_count>
<value>7.8639624293829910e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>606</sample_count>
<value>-6.5938476875607277e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.8865798640253987e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0378650426864624e-01</quality>
<le>6.9388854503631592e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>533</sample_count>
<value>-2.7381819728273915e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1326</sample_count>
<value>2.8352394260546955e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.1102230246251577e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>2</var>
<quality>5.0225412845611572e-01</quality>
<le>2.5000000000000000e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1791</sample_count>
<value>-1.0029134530916375e-02</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>68</sample_count>
<value>4.8411186971158215e-01</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.2204460492503178e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0205397605895996e-01</quality>
<le>2.3689103126525879e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1838</sample_count>
<value>-6.5406523866491815e-03</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>21</sample_count>
<value>7.3822507765214740e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.8865798640253987e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0392109155654907e-01</quality>
<le>2.2119045257568359e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1836</sample_count>
<value>1.2851614092359220e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>23</sample_count>
<value>-7.0930664915331199e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-1.7208456881689956e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0325584411621094e-01</quality>
<le>1.6267855167388916e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1811</sample_count>
<value>-1.4416639629038558e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>48</sample_count>
<value>4.8685109113945280e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>1.5543122344752168e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0334036350250244e-01</quality>
<le>1.4522725343704224e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1776</sample_count>
<value>2.1134121503482612e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>83</sample_count>
<value>-3.2737254080837347e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.2204460492503136e-16</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>0</var>
<quality>5.0246149301528931e-01</quality>
<le>9.0829002857208252e-01</le></_></splits></_>
<_>
<depth>1</depth>
<sample_count>1219</sample_count>
<value>-8.5346673521825803e-02</value>
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
<sample_count>640</sample_count>
<value>5.7886290686531831e-02</value>
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
<sample_count>1859</sample_count>
<value>4.4408920985006242e-16</value>
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
<quality>5.0454872846603394e-01</quality>
<le>7.8601944446563721e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>774</sample_count>
<value>1.8066758954429543e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1085</sample_count>
<value>-5.0945701357667621e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>3.1086244689504288e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
<_><var>0</var>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<quality>5.0212800502777100e-01</quality>
<le>1.5322914123535156e+00</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1796</sample_count>
<value>-1.4167158667728857e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>63</sample_count>
<value>3.0999551339753734e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-8.8817841970012602e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0266414880752563e-01</quality>
<le>1.0847599804401398e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>754</sample_count>
<value>2.2656461251734570e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1105</sample_count>
<value>-2.3970407856964623e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>8.8817841970012444e-16</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0370788574218750e-01</quality>
<le>1.0789400339126587e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>742</sample_count>
<value>-3.1940419009583935e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1117</sample_count>
<value>2.3657134249111228e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.2204460492503136e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0368273258209229e-01</quality>
<le>1.0177600383758545e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>590</sample_count>
<value>4.4494384607886871e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1269</sample_count>
<value>-1.6718599716109608e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>0.</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0391924381256104e-01</quality>
<le>1.0132449865341187e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>577</sample_count>
<value>-5.2562531380568607e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1282</sample_count>
<value>1.5038797351826086e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.1649348980190600e-15</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0355648994445801e-01</quality>
<le>9.2722505331039429e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>392</sample_count>
<value>7.2872834808621612e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1467</sample_count>
<value>-9.8094551697467572e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>6.6613381477509353e-16</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0229591131210327e-01</quality>
<le>1.4143149554729462e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1571</sample_count>
<value>-1.1560090798064190e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>288</sample_count>
<value>4.2036013274394679e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.2204460492503136e-16</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0365710258483887e-01</quality>
<le>1.0963299870491028e-01</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>781</sample_count>
<value>2.1235893234003624e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1078</sample_count>
<value>-3.5916730278930570e-01</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>-2.2204460492503178e-15</value>
<norm_class_idx>0</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0305408239364624e-01</quality>
<le>9.1055497527122498e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>376</sample_count>
<value>-7.5233897759775223e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1483</sample_count>
<value>8.1521687774440590e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_>
<_>
<best_tree_idx>-1</best_tree_idx>
<nodes>
<_>
<depth>0</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1859</sample_count>
<value>2.4424906541753385e-15</value>
<norm_class_idx>1</norm_class_idx>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error>
<splits>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<_><var>1</var>
<quality>5.0231134891510010e-01</quality>
<le>8.8118501007556915e-02</le></_></splits></_>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>322</sample_count>
<value>7.4755757692778257e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>1</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_>
<_>
<depth>1</depth>
better classifiers performance by training them with a new dataset of "manually" selected regions
11 years ago
<sample_count>1537</sample_count>
<value>-6.2031478956482825e-02</value>
Class-specific Extremal Region Filter algorithm as proposed in : Neumann L., Matas J.: Real-Time Scene Text Localization and Recognition, CVPR 2012. High-level C++ interface and implementation of algorithm is in the objdetect module. C++ example, a test image, and the default classifiers in xml files.
11 years ago
<norm_class_idx>0</norm_class_idx>
<Tn>0</Tn>
<complexity>0</complexity>
<alpha>0.</alpha>
<node_risk>0.</node_risk>
<tree_risk>0.</tree_risk>
<tree_error>0.</tree_error></_></nodes></_></trees></boost>
</opencv_storage>