Chiebot-Mirror
/
opencv_contrib
8
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

add training shell script in cnn_3dobj/samples/build

Browse Source
pull/276/head
Wangyida 10 years ago
parent 109da10f4f
commit 8aa47cac7e
10 changed files with 692 additions and 358 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. BIN
      modules/cnn_3dobj/samples/build/binary_label_test
  2. BIN
      modules/cnn_3dobj/samples/build/binary_label_train
  3. 113
      modules/cnn_3dobj/samples/build/linemod_triplet.prototxt
  4. 25
      modules/cnn_3dobj/samples/build/linemod_triplet_solver.prototxt
  5. 500
      modules/cnn_3dobj/samples/build/linemod_triplet_train_test.prototxt
  6. 36
      modules/cnn_3dobj/samples/build/train_linemod_triplet.sh
  7. 351
      modules/cnn_3dobj/samples/data/label_ape.txt
  8. 0
      modules/cnn_3dobj/samples/data/label_cat.txt
  9. 0
      modules/cnn_3dobj/samples/data/label_duck.txt
  10. 25
      modules/cnn_3dobj/samples/sphereview_3dobj_demo.cpp

BIN
modules/cnn_3dobj/samples/build/binary_label_test
View File

Binary file not shown.

BIN
modules/cnn_3dobj/samples/build/binary_label_train
View File

Binary file not shown.

113
modules/cnn_3dobj/samples/build/linemod_triplet.prototxt
Unescape View File

@ -0,0 +1,113 @@
name: "lfw_siamese"
input: "data"
input_dim: 10000
input_dim: 1
input_dim: 150
input_dim: 130
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 20
kernel_size: 5
stride: 1
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "conv1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 50
kernel_size: 5
stride: 1
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "conv2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "ip1"
type: "InnerProduct"
bottom: "pool2"
top: "ip1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 500
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "ip1"
top: "ip1"
}
layer {
name: "ip2"
type: "InnerProduct"
bottom: "ip1"
top: "ip2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 10
}
}
layer {
name: "feat"
type: "InnerProduct"
bottom: "ip2"
top: "feat"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 2
}
}

25
modules/cnn_3dobj/samples/build/linemod_triplet_solver.prototxt
Unescape View File

@ -0,0 +1,25 @@
# The train/test net protocol buffer definition
net: "examples/triplet/lfw_triplet_train_test.prototxt"
# test_iter specifies how many forward passes the test should carry out.
# In the case of lfw, we have test batch size 100 and 100 test iterations,
# covering the full 10,000 testing images.
test_iter: 100
# Carry out testing every 500 training iterations.
test_interval: 500
# The base learning rate, momentum and the weight decay of the network.
base_lr: 0.01
momentum: 0.9
weight_decay: 0.0000
# The learning rate policy
lr_policy: "inv"
gamma: 0.0001
power: 0.75
# Display every 100 iterations
display: 100
# The maximum number of iterations
max_iter: 50000
# snapshot intermediate results
snapshot: 5000
snapshot_prefix: "examples/triplet/lfw_triplet"
# solver mode: CPU or GPU
solver_mode: CPU

500
modules/cnn_3dobj/samples/build/linemod_triplet_train_test.prototxt
Unescape View File

@ -0,0 +1,500 @@
name: "lfw_triplet_train_test"
layer {
name: "triplet_data"
type: "Data"
top: "triplet_data"
top: "sim"
include {
phase: TRAIN
}
transform_param {
scale: 0.00390625
}
data_param {
source: "examples/triplet/lfw_triplet_train_leveldb"
batch_size: 64
}
}
layer {
name: "triplet_data"
type: "Data"
top: "triplet_data"
top: "sim"
include {
phase: TEST
}
transform_param {
scale: 0.00390625
}
data_param {
source: "examples/triplet/lfw_triplet_test_leveldb"
batch_size: 100
}
}
layer {
name: "slice_triplet"
type: "Slice"
bottom: "triplet_data"
top: "data"
top: "data_true"
top: "data_false"
slice_param {
slice_dim: 1
slice_point: 1
slice_point: 2
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
name: "conv1_w"
lr_mult: 1
}
param {
name: "conv1_b"
lr_mult: 2
}
convolution_param {
num_output: 20
kernel_size: 5
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "conv1"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
name: "conv2_w"
lr_mult: 1
}
param {
name: "conv2_b"
lr_mult: 2
}
convolution_param {
num_output: 50
kernel_size: 5
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "conv2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "ip1"
type: "InnerProduct"
bottom: "pool2"
top: "ip1"
param {
name: "ip1_w"
lr_mult: 1
}
param {
name: "ip1_b"
lr_mult: 2
}
inner_product_param {
num_output: 500
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "ip1"
top: "ip1"
}
layer {
name: "ip2"
type: "InnerProduct"
bottom: "ip1"
top: "ip2"
param {
name: "ip2_w"
lr_mult: 1
}
param {
name: "ip2_b"
lr_mult: 2
}
inner_product_param {
num_output: 10
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "feat"
type: "InnerProduct"
bottom: "ip2"
top: "feat"
param {
name: "feat_w"
lr_mult: 1
}
param {
name: "feat_b"
lr_mult: 2
}
inner_product_param {
num_output: 2
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "conv1_true"
type: "Convolution"
bottom: "data_true"
top: "conv1_true"
param {
name: "conv1_w"
lr_mult: 1
}
param {
name: "conv1_b"
lr_mult: 2
}
convolution_param {
num_output: 20
kernel_size: 5
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "pool1_true"
type: "Pooling"
bottom: "conv1_true"
top: "pool1_true"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "conv2_true"
type: "Convolution"
bottom: "pool1_true"
top: "conv2_true"
param {
name: "conv2_w"
lr_mult: 1
}
param {
name: "conv2_b"
lr_mult: 2
}
convolution_param {
num_output: 50
kernel_size: 5
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "pool2_true"
type: "Pooling"
bottom: "conv2_true"
top: "pool2_true"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "ip1_true"
type: "InnerProduct"
bottom: "pool2_true"
top: "ip1_true"
param {
name: "ip1_w"
lr_mult: 1
}
param {
name: "ip1_b"
lr_mult: 2
}
inner_product_param {
num_output: 500
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu1_true"
type: "ReLU"
bottom: "ip1_true"
top: "ip1_true"
}
layer {
name: "ip2_true"
type: "InnerProduct"
bottom: "ip1_true"
top: "ip2_true"
param {
name: "ip2_w"
lr_mult: 1
}
param {
name: "ip2_b"
lr_mult: 2
}
inner_product_param {
num_output: 10
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "feat_true"
type: "InnerProduct"
bottom: "ip2_true"
top: "feat_true"
param {
name: "feat_w"
lr_mult: 1
}
param {
name: "feat_b"
lr_mult: 2
}
inner_product_param {
num_output: 2
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "conv1_false"
type: "Convolution"
bottom: "data_false"
top: "conv1_false"
param {
name: "conv1_w"
lr_mult: 1
}
param {
name: "conv1_b"
lr_mult: 2
}
convolution_param {
num_output: 20
kernel_size: 5
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "pool1_false"
type: "Pooling"
bottom: "conv1_false"
top: "pool1_false"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "conv2_false"
type: "Convolution"
bottom: "pool1_false"
top: "conv2_false"
param {
name: "conv2_w"
lr_mult: 1
}
param {
name: "conv2_b"
lr_mult: 2
}
convolution_param {
num_output: 50
kernel_size: 5
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "pool2_false"
type: "Pooling"
bottom: "conv2_false"
top: "pool2_false"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "ip1_false"
type: "InnerProduct"
bottom: "pool2_false"
top: "ip1_false"
param {
name: "ip1_w"
lr_mult: 1
}
param {
name: "ip1_b"
lr_mult: 2
}
inner_product_param {
num_output: 500
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu1_false"
type: "ReLU"
bottom: "ip1_false"
top: "ip1_false"
}
layer {
name: "ip2_false"
type: "InnerProduct"
bottom: "ip1_false"
top: "ip2_false"
param {
name: "ip2_w"
lr_mult: 1
}
param {
name: "ip2_b"
lr_mult: 2
}
inner_product_param {
num_output: 10
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "feat_false"
type: "InnerProduct"
bottom: "ip2_false"
top: "feat_false"
param {
name: "feat_w"
lr_mult: 1
}
param {
name: "feat_b"
lr_mult: 2
}
inner_product_param {
num_output: 2
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "loss"
type: "TripletLoss"
bottom: "feat"
bottom: "feat_true"
bottom: "feat_false"
bottom: "sim"
top: "loss"
triplet_loss_param {
margin: 0.2
}
}

36
modules/cnn_3dobj/samples/build/train_linemod_triplet.sh
Unescape View File

@ -0,0 +1,36 @@
#!/usr/bin/env sh
# This script converts the lfw data into leveldb format.
git clone https://github.com/Wangyida/caffe/tree/cnn_triplet
cd caffe
mkdir build
cd build
cmake -DCMAKE_INSTALL_PREFIX=/usr/local ..
make -j4
make test
make install
cd ..
cmake ..
make -j4
./sphereview_test -radius=250 -ite_depth=4 -plymodel=../ape.ply -imagedir=../data/images_ape/ -labeldir=../data/label_ape.txt -num_class=3 -label_class=0
./sphereview_test -radius=250 -ite_depth=4 -plymodel=../duck.ply -imagedir=../data/images_duck/ -labeldir=../data/label_duck.txt -num_class=3 -label_class=1
./sphereview_test -radius=250 -ite_depth=4 -plymodel=../cat.ply -imagedir=../data/images_cat/ -labeldir=../data/label_cat.txt -num_class=3 -label_class=2
echo "Creating leveldb..."
rm -rf ./linemod_triplet_train_leveldb
rm -rf ./linemod_triplet_test_leveldb
convert_lfw_triplet_data \
./binary_image_train \
./binary_label_train \
./linemod_triplet_train_leveldb
convert_lfw_triplet_data \
./binary_image_test \
./binary_image_test \
./linemod_triplet_test_leveldb
echo "Done."
caffe train --solver=examples/triplet/lfw_triplet_solver.prototxt

351
modules/cnn_3dobj/samples/data/label_ape.txt
Unescape View File

@ -1,351 +0,0 @@
0 0.707107 0.707107
0.211325 0.57735 0.788675
-0.211325 0.57735 0.788675
0.408248 0.408248 0.816497
0 0.447214 0.894427
0.211325 0.57735 0.788675
-0.408248 0.408248 0.816497
-0.211325 0.57735 0.788675
0.211325 0.57735 0.788675
-0.211325 0.57735 0.788675
0.707107 0 0.707107
0.382683 0 0.92388
0.57735 0.211325 0.788675
0 0 1
0.214187 0.214187 0.953021
0.57735 0.211325 0.788675
0.57735 0.211325 0.788675
-0.707107 0 0.707107
-0.57735 0.211325 0.788675
-0.382683 0 0.92388
-0.214187 0.214187 0.953021
-0.57735 0.211325 0.788675
-0.57735 0.211325 0.788675
-0.707107 0.707107 0
-0.57735 0.788675 0.211325
-0.408248 0.816497 0.408248
-0.211325 0.788675 0.57735
0 0.707107 -0.707107
0 0.92388 -0.382683
0 1 4.21468e-08
0 0.92388 0.382683
0.211325 0.788675 -0.57735
0 0.92388 -0.382683
0.408248 0.816497 -0.408248
0.214186 0.953021 -0.214187
0 1 0
0.707107 0.707107 0
0.57735 0.788675 0.211325
0.57735 0.788675 -0.211325
0.408248 0.816497 0.408248
0.447214 0.894427 0
0 0.92388 0.382683
0.211325 0.788675 0.57735
0.214186 0.953021 0.214187
0.788675 0.57735 0.211325
0.816497 0.408248 0.408248
0.788675 0.211325 0.57735
0.707107 -0.707107 0
0.788675 -0.57735 0.211325
0.92388 -0.382683 0
0.816497 -0.408248 0.408248
0.953021 -0.214186 0.214187
1 4.21468e-08 0
0.788675 -0.211325 0.57735
0.894427 1.63234e-08 0.447214
0.92388 0.382683 0
0.953021 0.214187 0.214187
0.707107 0 -0.707107
0.788675 -0.211325 -0.57735
0.57735 0.211325 -0.788675
0.816497 -0.408248 -0.408248
0.408248 0.408248 -0.816497
0.57735 0.211325 -0.788675
0.57735 0.211325 -0.788675
0.816497 -0.408248 -0.408248
0.788675 -0.57735 -0.211325
0.211325 0.57735 -0.788675
0.408248 0.408248 -0.816497
0.788675 0.211325 -0.57735
0.816497 0.408248 -0.408248
0.639602 0.426401 -0.639602
0.788675 0.57735 -0.211325
0.408248 0.816497 -0.408248
0.639602 0.639602 -0.426401
0.211325 0.57735 -0.788675
0.211325 0.788675 -0.57735
0.426401 0.639602 -0.639602
0.211325 0.57735 -0.788675
0.211325 0.57735 -0.788675
-0.707107 0 -0.707107
-0.382683 0 -0.92388
0 0 -1
0.382683 0 -0.92388
0 -0.707107 -0.707107
0.211325 -0.57735 -0.788675
0.408248 -0.408248 -0.816497
0.211325 -0.57735 -0.788675
0.211325 -0.57735 -0.788675
0.57735 -0.211325 -0.788675
0.57735 -0.211325 -0.788675
0.57735 -0.211325 -0.788675
0.788675 -0.57735 -0.211325
0.57735 -0.788675 -0.211325
0.639602 -0.639602 -0.426401
0.408248 -0.816497 -0.408248
0.639602 -0.426401 -0.639602
0.211325 -0.788675 -0.57735
0.426401 -0.639602 -0.639602
0 -0.707107 0.707107
0.211325 -0.788675 0.57735
0 -0.92388 0.382683
0.408248 -0.816497 0.408248
0.214187 -0.953021 0.214186
0 -1 -4.21468e-08
0.57735 -0.788675 0.211325
0.447214 -0.894427 -1.63234e-08
0 -0.92388 -0.382683
0.214187 -0.953021 -0.214187
-0.707107 -0.707107 0
-0.57735 -0.788675 0.211325
-0.408248 -0.816497 0.408248
-0.211325 -0.788675 0.57735
-0.57735 -0.211325 0.788675
-0.408248 -0.408248 0.816497
-0.57735 -0.211325 0.788675
-0.57735 -0.211325 0.788675
-0.211325 -0.57735 0.788675
-0.211325 -0.57735 0.788675
-0.211325 -0.57735 0.788675
0.57735 -0.211325 0.788675
0.408248 -0.408248 0.816497
0.57735 -0.211325 0.788675
0.57735 -0.211325 0.788675
0.211325 -0.57735 0.788675
0.211325 -0.57735 0.788675
0.211325 -0.57735 0.788675
0.788675 -0.211325 0.57735
0.816497 -0.408248 0.408248
0.57735 0.211325 0.788675
-nan -nan -nan
-nan -nan -nan
-nan -nan -nan
-nan -nan -nan
-nan -nan -nan
-nan -nan -nan
-nan -nan -nan
-nan -nan -nan
06898 7.62273e-09 0.59069
0.788675 -0.211325 0.57735
0.874728 0.208721 0.437364
0.894427 1.63234e-08 0.447214
0.849313 0.106637 0.517007
0.809511 -0.31246 0.497052
0.874728 -0.208721 0.437364
0.849313 -0.106637 0.517007
0.83147 0.55557 0
0.92388 0.382683 0
0.867278 0.486184 0.10702
0.988185 0.108374 0.108374
0.980785 0.19509 0
0.953021 0.214187 0.214187
0.947388 0.301277 0.108113
0.895337 0.314938 0.314938
0.888679 0.403872 0.217112
0.935925 -0.108508 0.335068
0.935925 0.108508 0.335068
0.975663 2.28828e-08 0.219275
0.707107 0 -0.707107
0.754344 -0.106574 -0.64777
0.64777 0.106574 -0.754344
0.788675 -0.211325 -0.57735
0.57735 0.211325 -0.788675
0.816497 -0.408248 -0.408248
0.809511 -0.31246 -0.497052
0.408248 0.408248 -0.816497
0.497052 0.31246 -0.809511
0.57735 0.211325 -0.788675
0.57735 0.211325 -0.788675
0.64777 0.106574 -0.754344
0.707107 -7.71341e-09 -0.707107
0.788675 -0.57735 -0.211325
0.754344 -0.64777 -0.106574
0.816497 -0.408248 -0.408248
0.809511 -0.497052 -0.31246
0.788675 -0.57735 -0.211325
0.809511 -0.31246 -0.497052
0.809511 -0.497052 -0.31246
0.106574 0.64777 -0.754344
0.211325 0.57735 -0.788675
0.211325 0.57735 -0.788675
0.31246 0.497052 -0.809511
0.408248 0.408248 -0.816497
0.211325 0.57735 -0.788675
0.211325 0.57735 -0.788675
0.497052 0.31246 -0.809511
0.31246 0.497052 -0.809511
0.497052 0.31246 -0.809511
0.754344 0.106574 -0.64777
0.788675 0.211325 -0.57735
0.690768 0.213724 -0.690768
0.816497 0.408248 -0.408248
0.735924 0.421839 -0.529592
0.809511 0.31246 -0.497052
0.639602 0.426401 -0.639602
0.720687 0.321787 -0.614055
0.529592 0.421839 -0.735924
0.614055 0.321787 -0.720687
0.754344 0.64777 -0.106574
0.690768 0.690768 -0.213724
0.788675 0.57735 -0.211325
0.408248 0.816497 -0.408248
0.529592 0.735924 -0.421839
0.497052 0.809511 -0.31246
0.639602 0.639602 -0.426401
0.614055 0.720687 -0.321787
0.809511 0.497052 -0.31246
0.735924 0.529592 -0.421839
0.720687 0.614055 -0.321787
0.106574 0.64777 -0.754344
0.106574 0.754344 -0.64777
0.213724 0.690768 -0.690768
0.211325 0.788675 -0.57735
0.421839 0.529592 -0.735924
0.31246 0.497052 -0.809511
0.426401 0.639602 -0.639602
0.321787 0.614055 -0.720687
0.31246 0.809511 -0.497052
0.421839 0.735924 -0.529592
0.321787 0.720687 -0.614055
0.646997 0.539164 -0.539164
0.539164 0.646997 -0.539164
0.539164 0.539164 -0.646997
-0.707107 0 -0.707107
-0.55557 0 -0.83147
-0.382683 0 -0.92388
0 0 -1
-0.19509 0 -0.980785
0.55557 0 -0.83147
0.382683 0 -0.92388
0.19509 0 -0.980785
0 -0.707107 -0.707107
0.106574 -0.64777 -0.754344
0.211325 -0.57735 -0.788675
0.408248 -0.408248 -0.816497
0.31246 -0.497052 -0.809511
0.31246 -0.497052 -0.809511
0.211325 -0.57735 -0.788675
0.106574 -0.64777 -0.754344
0.64777 -0.106574 -0.754344
0.57735 -0.211325 -0.788675
0.57735 -0.211325 -0.788675
0.497052 -0.31246 -0.809511
0.754344 -0.64777 -0.106574
0.64777 -0.754344 -0.106574
0.690768 -0.690768 -0.213724
0.57735 -0.788675 -0.211325
0.735924 -0.529592 -0.421839
0.639602 -0.639602 -0.426401
0.720687 -0.614055 -0.321787
0.408248 -0.816497 -0.408248
0.497052 -0.809511 -0.31246
0.529592 -0.735924 -0.421839
0.614055 -0.720687 -0.321787
0.690768 -0.213724 -0.690768
0.529592 -0.421839 -0.735924
0.639602 -0.426401 -0.639602
0.614055 -0.321787 -0.720687
0.735924 -0.421839 -0.529592
0.720687 -0.321787 -0.614055
0.106574 -0.754344 -0.64777
0.211325 -0.788675 -0.57735
0.213724 -0.690768 -0.690768
0.421839 -0.735924 -0.529592
0.31246 -0.809511 -0.497052
0.426401 -0.639602 -0.639602
0.321787 -0.720687 -0.614055
0.421839 -0.529592 -0.735924
0.321787 -0.614055 -0.720687
0.646997 -0.539164 -0.539164
0.539164 -0.539164 -0.646997
0.539164 -0.646997 -0.539164
0 -0.707107 0.707107
0.106574 -0.754344 0.64777
0 -0.83147 0.55557
0.211325 -0.788675 0.57735
0.10702 -0.867278 0.486184
0 -0.92388 0.382683
0.408248 -0.816497 0.408248
0.314938 -0.895337 0.314938
0.31246 -0.809511 0.497052
0.214187 -0.953021 0.214186
0.217112 -0.888679 0.403872
0 -1 -4.21468e-08
0 -0.980785 0.19509
0.108374 -0.988185 0.108374
0.108113 -0.947388 0.301277
0.64777 -0.754344 0.106574
0.59069 -0.806898 -7.62273e-09
0.57735 -0.788675 0.211325
0.437364 -0.874728 -0.208721
0.447214 -0.894427 -1.63234e-08
0.517007 -0.849313 -0.106637
0.517007 -0.849313 -0.106637
0.517007 -0.849313 -0.106637
0.497052 -0.809511 0.31246
0.437364 -0.874728 0.208721
0.517007 -0.849313 0.106637
0.517007 -0.849313 0.106637
0.517007 -0.849313 0.106637
0 -0.83147 -0.55557
0 -0.92388 -0.382683
0.10702 -0.867278 -0.486184
0.108374 -0.988185 -0.108374
0 -0.980785 -0.19509
0.214187 -0.953021 -0.214187
0.108113 -0.947388 -0.301277
0.314938 -0.895337 -0.314938
0.217112 -0.888679 -0.403872
0.335068 -0.935925 0.108508
0.335068 -0.935925 -0.108508
0.219275 -0.975663 -2.28828e-08
-0.707107 -0.707107 0
-0.64777 -0.754344 0.106574
-0.57735 -0.788675 0.211325
-0.408248 -0.816497 0.408248
-0.497052 -0.809511 0.31246
-0.106574 -0.754344 0.64777
-0.211325 -0.788675 0.57735
-0.31246 -0.809511 0.497052
-0.64777 -0.106574 0.754344
-0.57735 -0.211325 0.788675
-0.408248 -0.408248 0.816497
-0.497052 -0.31246 0.809511
-0.497052 -0.31246 0.809511
-0.57735 -0.211325 0.788675
-0.64777 -0.106574 0.754344
-0.106574 -0.64777 0.754344
-0.211325 -0.57735 0.788675
-0.211325 -0.57735 0.788675
-0.31246 -0.497052 0.809511
0.64777 -0.106574 0.754344
0.57735 -0.211325 0.788675
0.57735 -0.211325 0.788675
0.408248 -0.408248 0.816497
0.497052 -0.31246 0.809511
0.106574 -0.64777 0.754344
0.211325 -0.57735 0.788675
0.31246 -0.497052 0.809511
0.211325 -0.57735 0.788675
0.106574 -0.64777 0.754344
0.31246 -0.497052 0.809511
0.754344 -0.106574 0.64777
0.57735 0.211325 0.788675
0.788675 -0.211325 0.57735
0.816497 -0.408248 0.408248
0.754344 -0.106574 0.64777
0.809511 -0.31246 0.497052
0.57735 0.211325 0.788675
0.211325 0.57735 0.788675
0.31246 0.497052 0.809511
0.497052 0.31246 0.809511

0
modules/cnn_3dobj/samples/data/label_cat.txt
Unescape View File

0
modules/cnn_3dobj/samples/data/label_duck.txt
Unescape View File

25
modules/cnn_3dobj/samples/sphereview_3dobj_demo.cpp
Unescape View File

@ -41,6 +41,10 @@
using namespace cv;
using namespace std;
using namespace cv::cnn_3dobj;
uint32_t swap_endian(uint32_t val) {
val = ((val << 8) & 0xFF00FF00) | ((val >> 8) & 0xFF00FF);
return (val << 16) | (val >> 16);
}
Point3d getCenter(string plymodel)
{
char* path_model=(char*)plymodel.data();
@ -79,13 +83,17 @@ void createHeader(int num_item, int rows, int cols, const char* headerPath)
std::ofstream headerImg(headerPathimg, ios::out|ios::binary);
std::ofstream headerLabel(headerPathlab, ios::out|ios::binary);
int headerimg[4] = {2051,num_item,rows,cols};
for (int i=0; i<4; i++)
headerimg[i] = swap_endian(headerimg[i]);
int headerlabel[2] = {2049,num_item};
for (int i=0; i<2; i++)
headerlabel[i] = swap_endian(headerlabel[i]);
headerImg.write(reinterpret_cast<const char*>(headerimg), sizeof(int)*4);
headerImg.close();
headerLabel.write(reinterpret_cast<const char*>(headerlabel), sizeof(int)*2);
headerLabel.close();
};
void writeBinaryfile(string filename, const char* binaryPath, const char* headerPath, int num_item)
void writeBinaryfile(string filename, const char* binaryPath, const char* headerPath, int num_item, int label_class)
{
int isrgb = 0;
cv::Mat ImgforBin = cv::imread(filename, isrgb);
@ -122,7 +130,7 @@ void writeBinaryfile(string filename, const char* binaryPath, const char* header
{
img_file.write(reinterpret_cast<const char*>(ImgforBin.ptr(r)), ImgforBin.cols*ImgforBin.elemSize());
}
unsigned char templab = 0;
unsigned char templab = (unsigned char)label_class;
lab_file << templab;
}
else
@ -136,19 +144,21 @@ void writeBinaryfile(string filename, const char* binaryPath, const char* header
{
img_file.write(reinterpret_cast<const char*>(ImgforBin.ptr(r)), ImgforBin.cols*ImgforBin.elemSize());
}
unsigned char templab = 0;
unsigned char templab = (unsigned char)label_class;
lab_file << templab;
}
img_file.close();
lab_file.close();
};
int main(int argc, char *argv[]){
const String keys = "{help | | demo :$ ./sphereview_test -radius=250 -ite_depth=1 -plymodel=../ape.ply -imagedir=../data/images_ape/ -labeldir=../data/label_ape.txt, then press 'q' to run the demo for images generation when you see the gray background and a coordinate.}"
const String keys = "{help | | demo :$ ./sphereview_test -radius=250 -ite_depth=2 -plymodel=../ape.ply -imagedir=../data/images_ape/ -labeldir=../data/label_ape.txt -num_class=2 -label_class=0, then press 'q' to run the demo for images generation when you see the gray background and a coordinate.}"
"{radius | 250 | Distanse from camera to object, used for adjust view for the reason that differet scale of .ply model.}"
"{ite_depth | 1 | Iteration of sphere generation, we add points on the middle of lines of sphere and adjust the radius suit for the original radius.}"
"{plymodel | ../ape.ply | path of the '.ply' file for image rendering. }"
"{imagedir | ../data/images_ape/ | path of the generated images for one particular .ply model. }"
"{labeldir | ../data/label_ape.txt | path of the generated images for one particular .ply model. }";
"{labeldir | ../data/label_ape.txt | path of the generated images for one particular .ply model. }"
"{num_class | 2 | total number of classes of models}"
"{label_class | 0 | class label of current .ply model}";
cv::CommandLineParser parser(argc, argv, keys);
parser.about("Demo for Sphere View data generation");
if (parser.has("help"))
@ -161,6 +171,8 @@ int main(int argc, char *argv[]){
string plymodel = parser.get<string>("plymodel");
string imagedir = parser.get<string>("imagedir");
string labeldir = parser.get<string>("labeldir");
int num_class = parser.get<int>("num_class");
int label_class = parser.get<int>("label_class");
cv::cnn_3dobj::IcoSphere ViewSphere(10,ite_depth);
std::vector<cv::Point3d> campos = ViewSphere.CameraPos;
std::fstream imglabel;
@ -180,7 +192,6 @@ int main(int argc, char *argv[]){
/// Let's assume camera has the following properties
Point3d cam_focal_point = getCenter(plymodel);
Point3d cam_y_dir(0.0f,0.0f,1.0f);
int num_obj = 1;
const char* headerPath = "./header_for_";
const char* binaryPath = "./binary_";
createHeader((int)campos.size(), 250, 250, headerPath);
@ -225,7 +236,7 @@ int main(int argc, char *argv[]){
filename = imagedir + filename;
filename += ".png";
myWindow.saveScreenshot(filename);
writeBinaryfile(filename, binaryPath, headerPath,(int)campos.size()*num_obj);
writeBinaryfile(filename, binaryPath, headerPath,(int)campos.size()*num_class, label_class);
}
return 1;
};

Write Preview
Loading…
Cancel
Save