ilya-lavrenov
a6084d2970
enabled some tests that failed before OCL_TEST_CYCLE*() macroses were defined and for some tests was disabled SANITY_CHECK, because plain and ocl results must not be completely equal
11 years ago
ilya-lavrenov
968c095403
gpu -> ocl
11 years ago
kdrobnyh
7a7aa148d4
Add IPP support in compare function
11 years ago
Jin Ma
114f3266d8
Removed performance test.
11 years ago
Jin Ma
4f3349ffe4
Added perf namespace.
11 years ago
Jin Ma
0233c4c198
Removed whitespace.
11 years ago
Jin Ma
14e083f1e0
Removed trailing whitespace
11 years ago
Jin Ma
5728612f95
Removed the trailing whitespace
11 years ago
yao
ab235cda74
fix warnings
11 years ago
yao
26b5eb3e39
add adaptive bilateral filter (cpp and ocl version)
11 years ago
Vadim Pisarevsky
f12702be24
fixed bugs in rectangular morphology case
11 years ago
Vadim Pisarevsky
771feb61c2
fixed building OpenCV with IPP 6.x (different DFT API); added Canny and rectangular morphology (based on Alex Kapustin patches)
11 years ago
Roman Donchenko
122d0f8685
Deleted the now-empty ocl perf_precomp.cpp.
...
See pull request #1241 for why.
11 years ago
ilya-lavrenov
3c86aa4ae1
appended ocl::finish to test_cycle defines
11 years ago
Vladislav Vinogradov
bff0fad6c3
gpu TVL1 Optical Flow optimization:
...
do not calculate sum of error in every round of iteration;
instead the error will be summed every 2nd times or more,
if the previous sum of error is too far away from threshold.
11 years ago
peng xiao
040b3b8265
Remove trailing whitespace.
11 years ago
peng xiao
c74c517ce3
Fix build error.
11 years ago
peng xiao
168d1dc18c
Make some tweaks to make `sum` operations less frequently
11 years ago
ilya-lavrenov
d02ccc9590
the code was refactored and old test system code was removed
11 years ago
ilya-lavrenov
4c28a6f0f6
most of the performance tests was rewritten in google-test manner
11 years ago
Vladislav Vinogradov
21c4753fed
used global memory access for up-scaling
11 years ago
Alexander Shishkov
88712d97c6
fix CMake changes
11 years ago
Vladislav Vinogradov
d1f6a23abf
improved texture usage:
...
* use explicit extrapolation only for sub-matrixes
* use built-in interpolation for INTER_NEAREST mode
11 years ago
Vladislav Vinogradov
241cc417f9
added custom implementation for nearest and linear interpolation
11 years ago
Vladislav Vinogradov
3b05acf936
reorganize code for further modifiction
11 years ago
Alexander Shishkov
396f6bb55f
fixed problems with building iOS version
11 years ago
MarkBelmont
cf34b3d65b
Changed the brief description of function Mat::colRange from "Create a matrix header for the specified row span." to "Create a matrix header for the specified column span."
11 years ago
Vladislav Vinogradov
f826bd8bce
removed NPP implementation
11 years ago
peng xiao
2519a21935
Fix a potential bug of ParallelLoopBodyWrapper::operator(Range)
...
On a 32-bit compiler the calculation may result in data (size_t) overflow when running some paralleled algorithms (which can safely run on a 64-bit compiler).
This bug is found when running OpenCV's Retina tutorial on 32bit VS2010.
11 years ago
Ilya Lavrenov
6c4ad9b597
fixer error with incorrect condition
11 years ago
Ilya Lavrenov
ec461a2ff0
added a test for ocl::norm
11 years ago
Jin Ma
1e8194fd3c
Optimized mog and mog2, which have much better performance.
11 years ago
Alexander Smorkalov
8aae54b7f4
WindowsRT mode build warning fix.
11 years ago
Roman Donchenko
3c137f7a04
Converted tabs to spaces.
11 years ago
Roman Donchenko
e9a28f66ee
Normalized file endings.
11 years ago
Roman Donchenko
f55740da70
Deleted all trailing whitespace.
11 years ago
Alexander Smorkalov
b15addec70
C4447 warning about main function format suppressed.
11 years ago
Alexander Smorkalov
d62c98527a
Invalid usage of cv::Ptr for arrays fixed.
11 years ago
Vladislav Vinogradov
de214950c4
minor gpu TVL1 optical flow optimization:
...
don't calc diff term if it is not used for epsilon criterion
11 years ago
Hanusz Leszek
32635a6834
using unsigned char instead of uint8_t to compile under windows
11 years ago
Vladislav Vinogradov
2899d558ac
record cuda gpu name to output report
11 years ago
Vladislav Vinogradov
b7ec673001
added cvconfig.h header, because ts_perf.cpp uses HAVE_CUDA macros
11 years ago
Hanusz Leszek
cf39ba5801
Allow to read PNG image of color_type PNG_COLOR_TYPE_PALETTE with alpha channel
...
Correct reading PNG color type palette with or without alpha imread flags -1 or 1
Better not using pnginfo.h, using png_get_tRNS instead
11 years ago
Vladislav Vinogradov
cb2985c526
fixed gpu bitwize operation:
...
incorrect types for template instantiation
11 years ago
Alexander Smorkalov
e37f7a4c73
Typo in Java generator fixed.
...
Typo breaks debug build for Android platform.
11 years ago
kdrobnyh
c1de14c27a
Fix bilateralFilter function
11 years ago
Vladislav Vinogradov
cfbd791465
fixed gpu cvtColor for BGR <-> YUV
11 years ago
kdrobnyh
3c9f307193
macros -> template in IPP support cvtColor
11 years ago
Nghia Ho
c6858c3fb0
Buffer is accessed out of bounds.
11 years ago
Roman Donchenko
7d768d2159
In calcOpticalFlowSF, fixed several uninitialized uses of matrices.
...
This should fix that pesky test failure that pops up from time to time.
I don't actually know if the default values should be zeros, but the tests
pass, so...
11 years ago