Andrey Pavlenko
bc741ab25f
Merge pull request #2009 from krodyush:pullreq/2.4-opt-131211-sepFilterSinglePass_final
2013-12-19 13:38:52 +04:00
Alexander Smorkalov
64c94cb22c
CUDA related func tables refactored to remove unneeded dependencies.
2013-12-19 11:18:04 +04:00
Alexander Smorkalov
6da7c50fb5
Make dependency from CUDA explicit to prevent from fake dependedcies from CUDA runtime.
2013-12-19 10:27:38 +04:00
Alexander Smorkalov
442082eb0f
Fixes for Android support.
2013-12-19 09:38:46 +04:00
Pierre-Emmanuel Viel
92fc763925
Fix some memory leaks in HierarchicalClusteringIndex
2013-12-18 22:46:41 +01:00
Andrey Pavlenko
459e7d4a80
Merge pull request #2015 from apavlenko:2.4_fix_ts_rng-fill
2013-12-18 19:13:22 +04:00
Andrey Pavlenko
a9687a341e
adding more than 4 channels random Mats support
...
if `Scalar:all` is used, Mats with 5+ channels cause errors
2013-12-18 17:27:39 +04:00
krodyush
1ae71fe205
intel device guard was added because of perf degradation on some non intel platform.
2013-12-18 14:27:51 +04:00
Alexander Smorkalov
be530bd085
DeviceInfo class method that were implemented in header moved to cpp file.
2013-12-18 12:02:15 +04:00
Vladimir Bystricky
80d0593dbd
Delete end comma in enumerations
2013-12-18 11:46:52 +04:00
Vladimir Bystricky
66145ea06c
Add CV_CAP_INTELPERC and CV_CAP_PROP_INTELPERC_ prefixes to const_ignore_list
2013-12-18 10:55:09 +04:00
krodyush
34c630faf4
update doc to be consisted with headers
2013-12-17 17:46:09 +04:00
Ilya Lavrenov
3a6d248bee
typo
2013-12-17 17:41:28 +04:00
krodyush
ea0c9b7f5c
GFFT fix for linux build
2013-12-17 17:12:57 +04:00
Vladimir Bystricky
cc08e00876
Fix notes about cmake files. Fix build warning.
2013-12-17 16:13:55 +04:00
krodyush
fffac2f085
optimize SURF by
...
Inlining and customizing sampling functions to reduce memory traffic and compute
Improve calcOrientation implementation.
Using more efficient rounding routines.
Removing unnecessary use of local memory
2013-12-17 14:12:33 +04:00
krodyush
dfe7c98090
optimize separable filter by
...
Added "sep_filter_singlepass" kernel that performs separable filtering in one kernel call
Added appropriate host part - sepFilter2D_SinglePass function and SingleStepSeparableFilterEngine_GPU class
Changed function declarations to enable their usage
2013-12-17 14:09:06 +04:00
krodyush
f3ee1c3d2f
Changes the datatype of the angle of the gradient for Intel platforms.
2013-12-17 14:06:14 +04:00
krodyush
917b883cf0
remove extra calculations from haar to be consistent with native implementation
2013-12-17 14:04:10 +04:00
krodyush
a63576e76d
HOST side optimization for GFFT
2013-12-17 14:02:57 +04:00
krodyush
de431609db
optimize Dx and Dy calcualtion to make it as single opencl kernel
2013-12-17 14:01:01 +04:00
Roman Donchenko
5d5527d03e
Merge pull request #1987 from pentschev:cv_completesymm_template_2.4
2013-12-17 13:17:25 +04:00
Vladimir Bystricky
4088013251
Add set/get depth generator properties by default. Add documentation
2013-12-17 12:00:40 +04:00
Alexander Smorkalov
88a883e68e
Build fix.
2013-12-17 10:24:00 +04:00
Alexander Smorkalov
8660e048bc
Dynamic CUDA support library loading implemented for Linux.
...
Logical mistake in macro fixed;
DeviceInfo deligate reimplemented;
Build and warning fixes.
2013-12-16 14:46:44 +04:00
Alexander Smorkalov
d4087f19a2
All CUDA related stuff were moved to separate dynamic library.
2013-12-16 14:46:44 +04:00
Peter Andreas Entschev
f44de302a0
cv::completeSymm fixed to work with any OpenCV data type and multiple channels.
2013-12-15 20:56:55 -02:00
Matthias Bady
b7c97f6e30
fixed init of clCxt
2013-12-15 22:41:07 +01:00
Vladimir Bystricky
e8d2a9752b
Add support Creative Senz3D camera by Intel Perceptual Computing SDK
2013-12-13 17:25:16 +04:00
Alexander Alekhin
17ca7f9545
ocl: add support for initialization from user-provided OpenCL handles
2013-12-12 19:59:22 +04:00
Alexander Alekhin
740b1f235e
ocl: implemented per-thread OpenCL command queue
2013-12-11 19:12:09 +04:00
Alexander Alekhin
a118577a57
core: added TLS support
2013-12-11 18:50:11 +04:00
Peter Andreas Entschev
c289c8a700
Fixing some indentation issues on FREAK source file.
2013-12-10 08:57:46 -02:00
Roman Donchenko
060e58d080
Merge pull request #1945 from apavlenko:2.4_version_update
2013-12-06 17:56:59 +04:00
Andrey Pavlenko
b5936648ca
Merge pull request #1943 from pentschev:ocl_oclMat_ptr_remove_2.4
2013-12-06 16:59:02 +04:00
Andrey Pavlenko
f2cf9dd8bf
updating version to 2.4.7.2
2013-12-06 16:21:22 +04:00
Andrey Pavlenko
7069b4b761
Merge pull request #1927 from JBosch:issue3409
2013-12-06 16:06:03 +04:00
Andrey Pavlenko
c34fc180a7
Merge pull request #1762 from jet47:cuda-6.0-npp-fix
2013-12-06 15:56:15 +04:00
Andrey Pavlenko
1d01c49a06
Merge pull request #1870 from es0m:es0m-2.4-imgproc-samplers
2013-12-06 15:47:39 +04:00
Peter Andreas Entschev
def17f2baa
Removing ocl::oclMat::ptr() method.
...
The method ocl::oclMat::ptr() is being removed because it returns a pointer to a row of oclMat::data. The data attribute is a cl_mem structure and cannot be iterated outside an OpenCL kernel.
2013-12-06 09:31:18 -02:00
Andrey Pavlenko
72cfff4c59
Merge pull request #1940 from pentschev:ocl_bfm_oclMat_ptr_fix_2.4
2013-12-06 15:05:34 +04:00
Andrey Pavlenko
b0b199ee8a
Merge pull request #1929 from alalek:ocl_haar_amd_beta
2013-12-06 14:56:47 +04:00
Peter Andreas Entschev
3bc952ed34
ocl::BruteForceMatcher fix wrong use of oclMat::ptr().
...
The oclMat::ptr() method was mistakenly used in ocl::BruteForceMatcher to pass a pointer to a oclMat object. The ptr() method returns a uchar pointer to the cl_mem data structure and this method will be removed.
2013-12-05 19:27:10 -02:00
Andrey Pavlenko
9dea590ada
Merge pull request #1933 from gongzg:fix-ocl-kernel-2.4
2013-12-05 17:05:25 +04:00
Alexander Alekhin
3dcddad88a
ocl: added workaround into Haar kernels
2013-12-05 16:56:01 +04:00
Andrey Pavlenko
39754bd6f9
Merge pull request #1920 from alalek:ocl_disable_workaround
2013-12-05 16:46:25 +04:00
Zhigang Gong
6ce03b0484
Fixed some implicitly type conversions between vector and scalar data type.
...
There are some mixed implicitly/explicitly type conversion between
scalar and vector. Although the spec allows those conversion, I prefer
to make them consistent and use explicitly all the cases.
Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
2013-12-05 09:35:21 +08:00
Josep Bosch
2153453fd1
Fixed little issue with mouse callback refering to issue 3409. Was using round instead of floor.
2013-12-04 22:32:20 +01:00
Ilya Lavrenov
eb7ff99f80
fixed cv::remap and cv::convertMaps for map types CV_16SC2 && CV_16UC1
2013-12-04 22:04:31 +04:00
Roman Donchenko
fa2cd93ecf
Merge pull request #1915 from jet47:gpu-color-test-fix
2013-12-04 13:55:28 +04:00
Alexander Alekhin
7b551af424
ocl: try to disable clFinish workaround
2013-12-04 13:00:37 +04:00
Andrey Pavlenko
8752414718
Merge pull request #1911 from apavlenko:2.4_fix_vc12
2013-12-03 19:20:05 +04:00
Andrey Pavlenko
1966e7cea9
fixing compilation on Linux
2013-12-03 17:46:09 +04:00
Alexander Alekhin
c962a9d61c
ocl:perf: fix moments test for plain impl
2013-12-03 16:02:55 +04:00
Vladislav Vinogradov
adb2040980
added additional check in cv::gpu::demosaicing that source is not empty
2013-12-03 15:55:47 +04:00
Vladislav Vinogradov
14ee306b9e
fix gpu test for Demosaicing:
...
check that input images was loaded correctly
2013-12-03 15:55:10 +04:00
Roman Donchenko
df136339ff
Merge pull request #1907 from severin-lemaignan:emscripten-2.4
2013-12-03 13:38:10 +04:00
Andrey Pavlenko
bd9d3dd561
fixing build of 2.4 branch with VC12 (aka 2013)
...
(master branch was fixed before, no need to merge this to master!)
- disable openexr (numerous undefined std::min & std::max)
- workaround for two compilator bugs
2013-12-03 13:16:44 +04:00
Séverin Lemaignan
9ce9fc3361
Ensure compilation to Javascript with Emscripten
2013-12-03 09:18:59 +01:00
Andrey Pavlenko
3ff1620086
Merge pull request #1899 from berak:b_3171_24
2013-12-03 11:32:31 +04:00
Andrey Pavlenko
8db6d9f9b2
Merge pull request #1906 from asmorkalov:android_ocl_warn_fix
2013-12-03 11:20:48 +04:00
Andrey Pavlenko
368bac7af6
Merge pull request #1744 from liuxing6666:2.4_fix_harris_test
2013-12-02 20:38:49 +04:00
Andrey Pavlenko
19ceab13f2
Merge pull request #1590 from ilya-lavrenov:cmd_parser_fix
2013-12-02 20:07:47 +04:00
Roman Donchenko
790f53f12a
Merge pull request #1887 from jet47:android-cuda-6.0
2013-12-02 18:53:26 +04:00
Alexander Smorkalov
eab7603096
Warning fixes and suppression in ocl module for gcc 4.6 from Android NDK.
2013-12-02 17:51:29 +04:00
Roman Donchenko
838961a289
Merge pull request #1897 from berak:b_3393_24
2013-12-02 17:11:56 +04:00
berak
34a390fbc2
CV_WRAP for BackgroundSubtractorMOG2
2013-11-30 13:35:23 +01:00
berak
059bf1d9e5
corrected initialization of output_folder in facereco examples
2013-11-30 13:09:42 +01:00
berak
773616ee04
swapped x,y in elbp_
2013-11-30 12:00:22 +01:00
Andrey Pavlenko
429ab8ebee
Merge pull request #1845 from StevenPuttemans:bugfix_3306
2013-11-29 23:08:20 +04:00
Alexander Smorkalov
a13db7c8fa
Android native camera destruction fixed for Android 4.4.
2013-11-29 10:38:26 +04:00
Vladislav Vinogradov
51663f5612
enable CUDA support for android
2013-11-28 10:00:47 +04:00
Vladislav Vinogradov
95ebdf4069
fixed compilation with CUDA 6.0:
...
GpuMat::setTo function, removed NPP call for CV_8S
2013-11-28 09:54:34 +04:00
Vladislav Vinogradov
7537521f53
disabled MOG2 performance test (sanity check fails on CUDA 6.0)
2013-11-28 09:50:05 +04:00
Vladislav Vinogradov
7c43daccba
fixed perf tests for bgfg segmentation
2013-11-28 09:49:50 +04:00
Roman Donchenko
10aec14a4d
Merge pull request #1849 from StevenPuttemans:feature_3375_documentation
2013-11-27 13:22:31 +04:00
Alexander Smorkalov
8fe4ab280b
Build error fix for binary-complience-checker tool.
2013-11-26 17:39:40 +04:00
Andrey Pavlenko
d45ce086c1
Merge pull request #1856 from alalek:ocl_fix_mt
2013-11-26 09:58:22 +04:00
Eric Sommerlade
3b4f65b186
fixed comparison to stop integer wrap around
2013-11-26 00:12:30 +00:00
StevenPuttemans
6f892b10ff
Applied feature #3375 and fix of pull 1842 to avoid merging problems later
2013-11-25 10:53:03 +01:00
StevenPuttemans
1b0297d92b
Added fixes mentioned in bugreport 3306
...
- Added author provided fixes
- Added code example
- Added documentation fixes
2013-11-25 09:35:23 +01:00
Roman Donchenko
4c5593b633
Merge pull request #1837 from asmorkalov:android_opencl
2013-11-25 12:19:20 +04:00
Roman Donchenko
06f6f74e61
Merge pull request #1839 from pmiecielica:memory_leaks
2013-11-25 12:17:08 +04:00
Roman Donchenko
59aa516ca5
Merge pull request #1860 from alalek:ocl_fix_undef_warn
2013-11-25 12:00:25 +04:00
Roman Donchenko
0537747f79
Merge pull request #1853 from StevenPuttemans:bugfix_3346
2013-11-22 18:26:16 +04:00
Alexander Alekhin
ddc9be9170
ocl: don't disable Undef warnings
2013-11-22 18:24:50 +04:00
Piotr Miecielica
a19cc6cb64
Added pointer initialization
2013-11-22 13:33:33 +01:00
Alexander Smorkalov
514e97223c
Some hacks to build and run OpenCL on Qualcomm S800.
...
Some of functions were enabled on Qualcomm S800 by changing grid size;
OpenCL kernel grid size unification for different platfroms;
Test pass rate improvements by inclreasing threshold;
Some tests were disabled for Android;
run.py was adopted for devices with brackets in in name.
2013-11-22 16:13:38 +04:00
Alexander Alekhin
d650efc069
ocl: multi-threading: opencl buffer memory guard
2013-11-22 14:39:51 +04:00
Alexander Alekhin
96a4a7b339
ocl: multi-threading: fix bug in intialization
2013-11-22 14:39:33 +04:00
StevenPuttemans
7a8bd31657
Fixed bug in report #3346
2013-11-22 09:57:47 +01:00
Piotr Miecielica
923698d700
Fixes of some memory leaks
2013-11-21 18:55:09 +01:00
Andrey Pavlenko
8a4f35704a
Merge pull request #1840 from ilya-lavrenov:ocl_cvtColor
2013-11-21 17:32:53 +04:00
Ilya Lavrenov
d764af5318
added extra parameters to ocl::cvtColor performance test
2013-11-21 12:53:21 +04:00
Roman Donchenko
6b23cb5b02
Merge pull request #1795 from alalek:perf_report_progress
2013-11-21 12:31:36 +04:00
Roman Donchenko
b8f5fcf3f9
Merge pull request #1776 from philippefoubert:merge_upstream_2.4_ximea
2013-11-21 12:03:02 +04:00
Roman Donchenko
623a02c69f
Merge pull request #1834 from SpecLad:svm-copy-ctor
2013-11-21 11:55:20 +04:00
Alexander Alekhin
c1541c9e79
perf report: summary.py: added --progress mode
2013-11-20 19:57:20 +04:00
Ilya Lavrenov
60b59eeb73
workaround for AMD bug: UNREACHABLE EXECUTED
2013-11-20 15:40:56 +04:00
Roman Donchenko
cf00349b5b
Gave CvSVM a private copy constructor & assignment operator.
...
The implicitly generated ones don't work properly, and I don't want to
write proper ones. :-)
"Fixes" <http://code.opencv.org/issues/3358 >.
2013-11-20 14:59:42 +04:00
Roman Donchenko
b197ec9413
Merge pull request #1761 from bhuroc:qtkit-video-file-2.4
2013-11-20 13:36:37 +04:00
Philippe FOUBERT
78e16a906b
Back to the previous coding way (using the macro with the goto).
2013-11-19 21:51:47 +01:00
Roman Donchenko
e69d2c1bb1
Merge pull request #1819 from ilya-lavrenov:ocl_resize_AREA
2013-11-19 16:25:47 +04:00
Roman Donchenko
c47190ef90
Merge pull request #1787 from hs3180:2.4
2013-11-19 14:21:25 +04:00
Ilya Lavrenov
198cd1a40d
added area fast mode to ocl::resize
2013-11-19 13:43:50 +04:00
Roman Donchenko
4746a710bb
Merge pull request #1820 from asmorkalov:android_native_camera_44
2013-11-19 12:47:48 +04:00
Roman Donchenko
3e55d705cc
Merge pull request #1823 from ilya-lavrenov:ocl_abs_sum
2013-11-19 12:34:09 +04:00
Roman Donchenko
56f791679b
Merge pull request #1825 from jet47:fix-opengl-mem-leak
2013-11-19 12:31:53 +04:00
Vladislav Vinogradov
66221a32b9
fixed memory leak in ogl::Texture2D
2013-11-19 10:12:22 +04:00
Ilya Lavrenov
c0456df611
workaround for ocl::absSum (eliminated error: unreachable executed)
2013-11-18 20:38:39 +04:00
Ilya Lavrenov
f70d63e4c9
added INTER_AREA interpolation to ocl::resize
2013-11-18 18:28:49 +04:00
Alexander Smorkalov
04d293d2c4
Initial NativeCamera support for Android 4.4 KitKat.
2013-11-18 18:07:26 +04:00
Roman Donchenko
cdf4e4e0b5
Merge pull request #1810 from Nerei:2.4
2013-11-18 17:49:40 +04:00
Roman Donchenko
4fd1960425
Merge pull request #1807 from KonstantinMatskevich:intel_bug
2013-11-18 15:16:37 +04:00
Roman Donchenko
3f927abb4d
Merge pull request #1809 from ilya-lavrenov:ocl_resize_nn
2013-11-18 14:55:24 +04:00
Konstantin Matskevich
327d9f274b
fixed Split for device version equals 78712
2013-11-18 13:20:57 +04:00
Konstantin Matskevich
0e1717c14c
fixing bugs for Intel platform CPU device
2013-11-18 13:20:54 +04:00
Roman Donchenko
f5ded41016
Merge pull request #1808 from alalek:ocl_fix_cvtcolor_xyz
2013-11-18 12:20:58 +04:00
Anatoly Baksheev
13bb236ad5
fixed some warnings
2013-11-16 21:18:36 +04:00
Anatoly Baksheev
8f6f9c3e58
compilation for VS2008
2013-11-16 20:14:01 +04:00
Alexander Alekhin
0a698bf145
ocl: fix cvtColor XYZ: variables on stack
2013-11-16 13:45:39 +04:00
Ilya Lavrenov
cc237b7aba
generalized ocl::resize for all data types (INTER_NEAREST mode)
2013-11-15 22:52:29 +04:00
Roman Donchenko
4e2ec7b97d
Merge pull request #1803 from ilya-lavrenov:ocl_cvtColor
2013-11-15 15:55:29 +04:00
Roman Donchenko
2e79dde357
Merge pull request #1799 from SpecLad:avformat-53-6
2013-11-15 11:51:59 +04:00
Roman Donchenko
3b7acf2657
Fix building with FFmpeg 0.8.
...
avformat_find_stream_info was introduced in Libav avformat 53.3,
but it was only merged to FFmpeg in avformat 53.6.
In Libav avformat 53.3-53.5 av_find_stream_info is not removed
(only deprecated), so this shouldn't break building with that.
2013-11-14 17:25:46 +04:00
Roman Donchenko
be88b20154
Merge pull request #1796 from ilya-lavrenov:ocl_cvtColor
2013-11-14 14:46:43 +04:00
Roman Donchenko
84da6ca265
Merge pull request #1792 from ilya-lavrenov:ocl_warp_eps
2013-11-14 14:46:06 +04:00
Roman Donchenko
19b3064706
Merge pull request #1790 from ilya-lavrenov:ocl_ref
2013-11-14 14:45:50 +04:00
Ilya Lavrenov
fe76b2116a
added RGBA <-> mRGBA
2013-11-14 12:02:59 +04:00
Ilya Lavrenov
1b7c5b201d
added HLS -> RGB[A][FULL] conversion
2013-11-13 19:35:24 +04:00
Ilya Lavrenov
98915e06bc
added HSV -> RGB[A][FULL] conversion
2013-11-13 19:10:06 +04:00
Ilya Lavrenov
3bdd9626f3
added RGB[A] -> HLS[FULL] conversion
2013-11-13 17:41:40 +04:00
Roman Donchenko
15f4292a1f
Merge pull request #1788 from alalek:perf_test_skip
2013-11-13 17:21:52 +04:00
Ilya Lavrenov
af7c614438
added RGB[A] -> HSV[FULL] conversion
2013-11-13 17:09:05 +04:00
Roman Donchenko
411629a57d
Merge pull request #1786 from SpecLad:cap-images-file-pattern
2013-11-13 17:00:20 +04:00
Roman Donchenko
54ea5bbac7
Merge pull request #1779 from perping:integral_2.4
2013-11-13 15:53:42 +04:00
Roman Donchenko
28e0d3d771
Merge pull request #1784 from SpecLad:ffmpeg-no-png
2013-11-13 15:52:42 +04:00
Alexander Alekhin
fb326a58fc
ocl: perf tests: skip tests if we fail check for maxMemoryAllocSize
2013-11-13 14:17:57 +04:00
Alexander Alekhin
f1873bbca1
perf tests: allow to skip performance tests
2013-11-13 14:17:56 +04:00
Ilya Lavrenov
6770c04073
refactored some functions from ocl arithm
2013-11-13 13:56:49 +04:00
Ilya Lavrenov
c38e4f9639
added eps to ocl::buildWarpPerspectiveMaps perf test
2013-11-13 13:04:02 +04:00
Roman Donchenko
376993be4c
Merge pull request #1785 from ilya-lavrenov:ocl_kern_warn
2013-11-13 12:20:09 +04:00
perping
185059959b
fix haar
2013-11-13 10:19:09 +08:00
perping
1e2fb6de30
fix match_template and haar
2013-11-13 10:09:39 +08:00
Ilya Lavrenov
0bf7350615
fixed compilation of ocl::minMaxLoc for Intel device
2013-11-12 19:14:40 +04:00
huangs
f02ce90559
minX1 should be max(-maxD,0) not max(maxD,0), avoid black band at left
2013-11-12 22:37:23 +08:00
Roman Donchenko
b8e3d3f791
In the image sequence capture, only search for the ordinal in the file name.
...
Searching in directory names can yield confusing results; e.g. if
the input is "jpeg2000/image1.jp2", it will infer the pattern
"jpeg%04d/image1.jp2", which is likely not what the user intended.
If the user really desires for the variable part to be in the
directory name, it can always use an explicit pattern.
2013-11-12 16:26:21 +04:00
Ilya Lavrenov
96121a66c5
kernel warnings on AMD
2013-11-12 15:06:46 +04:00
Roman Donchenko
a81efdbb25
Merge pull request #1775 from ilya-lavrenov:ocl_cvtColor
2013-11-12 14:26:50 +04:00
perping
d6e0ef2710
fix linux warning.
2013-11-12 18:13:44 +08:00
