Commit Graph

448 Commits

Author SHA1 Message Date
Alexey Spizhevoy
2f8af6335e renamed GpuFeature into FeatureSet and updated docs 2011-02-15 13:56:59 +00:00
Alexey Spizhevoy
04709a2793 refactoring of GPU module 2011-02-15 13:25:24 +00:00
Anatoly Baksheev
6b6a63ba38 fixed some warnings
GPU: TargetArchs -> added FEATURE_SET prefix.
2011-02-15 10:54:49 +00:00
Alexey Spizhevoy
725d83b0e5 added init() function into MultiGpuMgr, added samples 2011-02-15 10:03:26 +00:00
Alexey Spizhevoy
5d4913a2ee fixed multi GPU API, added driver_api prefix to multi gpu samples 2011-02-15 09:12:48 +00:00
Vladislav Vinogradov
0821c7ad17 fixed descriptor calculation in SURF_GPU 2011-02-15 08:57:35 +00:00
Vladislav Vinogradov
deac5d972e fixed errors in gpu on old video cards (SURF_GPU, BruteForceMatcher_GPU, min/max, setTo, convertTo)
added assertion after all kernels calls
2011-02-14 15:50:17 +00:00
Alexey Spizhevoy
a42a42858c fixed GPU samples and MultiGpuMgr 2011-02-14 15:09:45 +00:00
Anatoly Baksheev
d7e612cd4b GPU: In strcut _scanElemOp dummy type pass by value 2011-02-14 13:30:00 +00:00
Alexey Spizhevoy
186e46fcca added draft version of MultiGpuMgr (it isn't tested on multi GPU machine yet) 2011-02-14 12:53:59 +00:00
Alexey Spizhevoy
a6aff1856b removed check disabling GPU face detection under Linux, i've checked -- it works 2011-02-10 13:01:36 +00:00
Alexey Spizhevoy
63806c9ab9 renamed gpu::DeviceInfo::has into gpu::DeviceInfo::supports 2011-02-09 12:31:05 +00:00
Vladislav Vinogradov
924670d32c fixed block size calculation in SURF_GPU (fasthessian_gpu and nonmaxonly_gpu kernels) 2011-02-09 09:11:11 +00:00
Anatoly Baksheev
7539b7de65 fixed build under ubuntu, but FS is still disabled 2011-02-07 13:47:10 +00:00
Vladislav Vinogradov
5cd06d6a36 fixed SURF_GPU (fails on empty data)
added test for SURF_GPU and reprojectImageTo3D
2011-02-07 10:12:04 +00:00
Anatoly Baksheev
f42a449df9 fix compilation under win32 with gpu 2011-02-07 09:18:11 +00:00
Anatoly Baksheev
7b1c265563 fixed some compilation under ubuntu 2011-02-04 18:29:05 +00:00
Anatoly Baksheev
21f0d1e174 removed MS specific from FD code 2011-02-04 15:46:35 +00:00
Anatoly Baksheev
0747f2d863 1) NPP_staging as sources. Binaries removed.
2) NVidia tests for GPU
3) FD sample that uses NVidia's interface.
2011-02-04 15:15:25 +00:00
Alexey Spizhevoy
50429d8a3e fixed some warnings and errors under g++ 2011-02-03 14:51:58 +00:00
Alexey Spizhevoy
ea94b43541 added stereo_multi_gpu sample, cosmetic changes in multi_gpu sample 2011-02-03 12:02:39 +00:00
Alexey Spizhevoy
557dd39f03 fixed gpu::sum* on CC1.0, updated some tests 2011-02-02 07:23:55 +00:00
Alexey Spizhevoy
e303b0dd62 added CUDA_ARCH_FEATUERS define (for correct handling of BIN(PTX) cases) 2011-02-01 12:28:39 +00:00
Alexey Spizhevoy
16e74ab306 added buffered version of norm, updated performance sample and docs 2011-02-01 10:46:19 +00:00
Alexey Spizhevoy
3795142604 implemented gpu::norm via absSum, sqrSum, and minMax (removed norm_diff call), added support of other data types 2011-02-01 10:23:10 +00:00
Alexey Spizhevoy
ae529f4bc6 added absSum function 2011-01-31 14:37:03 +00:00
Alexey Spizhevoy
fa446e7e35 removed linear_filters_beta.cu as its functionality was moved into filters.cu 2011-01-31 13:31:59 +00:00
Vladislav Vinogradov
8274ed22e4 fixed gpu tests (BruteForceMatcher_GPU, divide, phase, cartToPolar, async)
minor code refactoring
2011-01-31 13:20:52 +00:00
Alexey Spizhevoy
7a29d96cf4 added buffered version of gpu::integral function and updated performance test (it still works too slow) 2011-01-31 10:42:33 +00:00
Alexey Spizhevoy
0af5356dbc updated code in case of the compilation without CUDA 2011-01-28 12:30:08 +00:00
Alexey Spizhevoy
575fd1fe4c reafactoring: replaced query device props functions with the DeviceInfo class 2011-01-28 11:59:26 +00:00
Alexey Spizhevoy
65b9f3bc10 fixed TargetArchs implementation in case when HAVE_CUDA=false, added initial structure for multi_gpu sample 2011-01-27 12:17:56 +00:00
Alexey Spizhevoy
891e2ff310 replaced has* methods in the GPU module with the TargetArchs monostate 2011-01-27 10:06:38 +00:00
Alexey Spizhevoy
13a6d0b92a fixed parsing GPU archs in BIN(PTX) format 2011-01-27 08:26:10 +00:00
Vladislav Vinogradov
eda8416358 fixed BruteForceMatcher_GPU (fails if input data is empty)
updated BruteForceMatcher_GPU test
2011-01-26 15:58:47 +00:00
Alexey Spizhevoy
310c483da8 added first version of gpu performance tests 2011-01-24 10:33:01 +00:00
Vladislav Vinogradov
811ba31897 added synchronization after NPP calls 2011-01-24 10:32:57 +00:00
Vladislav Vinogradov
8abdb3721f added gpu threshold. 2011-01-24 10:11:02 +00:00
Alexey Spizhevoy
055c226392 fixed FindNPP.cmake Apple-aimed part 2011-01-21 12:57:40 +00:00
Alexey Spizhevoy
e51b9021bc forgot to rename namespace 2011-01-21 11:31:45 +00:00
Alexey Spizhevoy
db41449be8 fixed errors under MacOS 2011-01-21 10:53:07 +00:00
Anatoly Baksheev
e3f3de84db warning fixed 2011-01-21 09:00:19 +00:00
Alexey Spizhevoy
01dafce1a1 fixed some bugs in GPU matrix reductions, removed <functional> into precomp.hpp 2011-01-21 07:43:11 +00:00
Alexey Spizhevoy
0da71a01ff fixed some GPU tests failing when compiled for 1.1(no doubles) and run on 1.3(with doubles) 2011-01-20 15:08:48 +00:00
Alexey Spizhevoy
9e48f64149 fixed parsing of CC in gpu module 2011-01-20 14:34:27 +00:00
Alexey Spizhevoy
574b3f94a1 updated gpu initialization functions, added compile-time error on CC 1.0 2011-01-20 14:13:07 +00:00
Alexey Spizhevoy
8779306800 updated main CMakeLists.txt gpu module section, now user can manage binary and intermediate code versions of the gpu module image
added more functions to check version of gpu code in runtime
2011-01-20 09:22:05 +00:00
Alexey Spizhevoy
4b8425dbb8 fixed minor bugs 2011-01-19 13:28:22 +00:00
Alexey Spizhevoy
186b1fc6ef refactored gpu module 2011-01-19 12:47:34 +00:00
Alexey Spizhevoy
90ae1e3aed refactored gpu module 2011-01-19 10:54:58 +00:00
Alexey Spizhevoy
1a0d41fb53 added checkPtxVersion into gpu module 2011-01-18 14:52:35 +00:00
Alexey Spizhevoy
cbb132ccb1 added ensureSizeIsEnough into gpu module, updated reduction methods 2011-01-18 12:36:01 +00:00
Alexey Spizhevoy
f3a2656808 added hasPtxFor and isCompatibleWith functions into gpu module, added docs for them 2011-01-18 12:01:28 +00:00
Anatoly Baksheev
cb63046dcf gpu face detetcion:
1) fixed bug with error codes (enum NppStStatus) shift.
2) added some asserts
2011-01-17 11:32:38 +00:00
Alexey Spizhevoy
db852e0b54 updated gpu module docs 2011-01-13 13:55:13 +00:00
Vladislav Vinogradov
349e0ece93 added docs for GPU Filter Engine 2011-01-13 13:48:58 +00:00
Anatoly Baksheev
1a94186195 First version of CascadeClassifier_GPU.
Only for VS2008 now.
Sample for it.
new NPP_staging for VS2008 only
2011-01-13 13:04:00 +00:00
Alexey Spizhevoy
4e23f37ff8 added docs for gpu::HOGDescriptor 2011-01-12 13:07:33 +00:00
Alexey Spizhevoy
4dfbf99dd5 cosmetic changes in gpu module, decreased matchTemplate test running time 2011-01-12 09:30:08 +00:00
Alexey Spizhevoy
d091ae5746 removed unnecessary code from gpu::matchTemplate 2011-01-12 06:49:03 +00:00
Alexey Spizhevoy
57f917d6f2 replaced adding constant in normalization with taking max 2011-01-11 10:15:46 +00:00
Alexey Spizhevoy
dc763e0250 updated normalization routine in the matchTemplate to avoid division by zero on black images (ticket #798), added test 2011-01-11 09:36:21 +00:00
Alexey Spizhevoy
2f13e4ce58 refactored hog, added camera support into hog_sample 2010-12-29 15:45:01 +00:00
Alexey Spizhevoy
ba32833c3f added missed include (for MSVC 2010) 2010-12-28 14:57:23 +00:00
Alexey Spizhevoy
be38864dd0 added buffered version of gpu::convolve 2010-12-27 10:18:42 +00:00
Alexey Spizhevoy
1ecb6cf775 simplified gpu::columnSum test, it doesn't fail on Quadro anymore (when seed is 000001af5a11badd) after BFM test, but something definitely wrong with NPP_Staging's transpose 2010-12-27 09:10:22 +00:00
Alexey Spizhevoy
8f0d36b8b6 refactored gpu::dft 2010-12-27 07:35:41 +00:00
Alexey Spizhevoy
86802ec968 minor warning fixed 2010-12-24 13:44:26 +00:00
Alexey Spizhevoy
eaf35a8421 minor refactoring in gpu module 2010-12-24 12:55:43 +00:00
Alexey Spizhevoy
21b081deff now single row GPU matrix is continuous one, added aux. functions, updated dft and matchTemplates 2010-12-24 09:26:19 +00:00
Alexey Spizhevoy
54fcdf4cae minor changes in gpu module (comments and warnings) 2010-12-24 07:00:08 +00:00
Alexey Spizhevoy
6702d55711 added support of scaling into gpu::dft, refactored gpu::convolve 2010-12-24 06:48:23 +00:00
Alexey Spizhevoy
783716838d updated test for gpu::dft, updated dft for handling continous source 2010-12-23 13:00:33 +00:00
Alexey Spizhevoy
52ca0c4bca fixed bug in gpu::dft 2010-12-23 10:07:37 +00:00
Alexey Spizhevoy
09735fd208 added gpu::dft implemented via CUFFT 2010-12-23 09:24:33 +00:00
Alexey Spizhevoy
68aba9f2fb added mulSpectrums functions into GPU module 2010-12-22 13:46:06 +00:00
Alexey Spizhevoy
fef06c25b5 moved crossCorr (as NPP_Staging wrapper) into public GPU module part from the internal matchTemplate files 2010-12-22 08:56:16 +00:00
Alexey Spizhevoy
f9bcef9003 moved sqrIntegral (NPP_Staging wrapper) into public GPU module part from matchTemplate.cpp 2010-12-22 08:17:36 +00:00
Alexey Spizhevoy
a6d9cce500 added CUFFT errors handling into GPU module 2010-12-22 08:03:53 +00:00
Vladislav Vinogradov
8190837dd4 fixed some warnings under linux 2010-12-22 07:30:21 +00:00
Vladislav Vinogradov
0cd587ee34 added gpu transpose and integral based on NPP Staging.
added mask support to SURF_GPU.
2010-12-21 14:02:09 +00:00
Vladislav Vinogradov
5bfb44f887 fixed compiler error under linux 2010-12-21 08:13:15 +00:00
Vladislav Vinogradov
0e45a637c4 added SURF_GPU.
added support of CV_32FC1 type to gpu copyMakeBorder.
2010-12-20 12:49:40 +00:00
Alexey Spizhevoy
df8529377b refactoring: moved gpu reduction-based functions into separated file 2010-12-20 09:51:25 +00:00
Alexey Spizhevoy
1922e50f19 refactoring: made gpu bitwise operations inline 2010-12-20 09:16:17 +00:00
Alexey Spizhevoy
0465b89e7e gpu module refactoring: moved per-element operations into separated file 2010-12-20 09:07:19 +00:00
Alexey Spizhevoy
6891a60149 added host code for gpu::matchTemplate (as NPP_staging was integrated) 2010-12-20 08:09:09 +00:00
Alexey Spizhevoy
e62bf3a2ae updated gpu bitwise operations 2010-12-20 08:06:13 +00:00
Anatoly Baksheev
0f30fe080f renamed constantspacebp -> stereocsbp line all other algs
created element_operations and matrix_reductions files for farther refactoring
2010-12-19 17:20:54 +00:00
Anatoly Baksheev
9dd4a22a5e temporary added NPP_staging, functionality from the library will be moved to NPP with next release. 2010-12-17 15:41:26 +00:00
Alexey Spizhevoy
7767038ef0 updated other gpu's bitwise operations 2010-12-17 12:48:04 +00:00
Alexey Spizhevoy
5132ce211b reimplemented gpu::bitwise_not operation, refactored gpu module 2010-12-17 12:22:51 +00:00
Alexey Spizhevoy
1a93412eca refactored bitwise operations in gpu module 2010-12-17 10:26:57 +00:00
Alexey Spizhevoy
74197c5b14 fixed link time error in gpu module 2010-12-16 08:46:47 +00:00
Alexey Spizhevoy
343c33d73e added support of CCOEFF_NORMED for multichannel images (8U) into gpu::matchTemplate 2010-12-16 08:10:31 +00:00
Alexey Spizhevoy
640af6623c added support of multichannel images into gpu::sqrSum 2010-12-15 16:32:56 +00:00
Alexey Spizhevoy
3db5b687f6 added support of multichannel images into gpu::matchTemplate for CCOEFF method 2010-12-15 16:04:10 +00:00
Alexey Spizhevoy
f56d9c340f added support of remaining image number of channels into gpu::sum 2010-12-15 15:28:35 +00:00
Alexey Spizhevoy
d8a7ff1e00 refactored gpu module, added vec math operators for uint, added support of 2 channel images into gpu::sum (removed support of double) 2010-12-15 15:12:32 +00:00
Alexey Spizhevoy
e5eec31be1 fixed minor bugs in gpu module 2010-12-15 12:10:30 +00:00
Alexey Spizhevoy
68c3018047 added support of multichannel images into gpu::matchTemplate (all methods except CCOEFF based), refactored 2010-12-15 11:22:37 +00:00
Alexey Spizhevoy
e7cf541f5f fixed bug in matchTemplate when template size is (1,1), refactored 2010-12-14 09:53:17 +00:00
Alexey Spizhevoy
1887b7d2e4 refactored matchTemplate.cu 2010-12-14 08:45:11 +00:00
Alexey Spizhevoy
ce47a37e6e added CUDA kernel for CV_TM_CCOEFF_NORMED matchTemplate method 2010-12-14 08:00:53 +00:00
Alexey Spizhevoy
b35aa77418 added CUDA kernel for CV_TM_CCOEFF matchTemplate method 2010-12-14 07:42:55 +00:00
Alexey Spizhevoy
39700c5d54 added some gpu::matchTemplate kernels (other parts after NPP Staging integration) 2010-12-13 16:48:34 +00:00
Alexey Spizhevoy
a81b41fb08 fixed some warning under Ubuntu in gpu module 2010-12-13 15:56:29 +00:00
Alexey Spizhevoy
c9f9f38777 added gpu::sqrSum function 2010-12-13 14:34:02 +00:00
Vladislav Vinogradov
b18a3a5f83 fixed errors in StereoBeliefPropogation under linux 2010-12-13 13:52:40 +00:00
Anatoly Baksheev
070d87fb7f fixed bug with prefilter_xsobel preset. 2010-12-13 13:42:35 +00:00
Alexey Spizhevoy
e193fa1165 fixed minor bug in gpu::sum 2010-12-13 12:50:52 +00:00
Alexey Spizhevoy
3997514b7c added tests for gpu::sum, it supports all data types, but single channel images only 2010-12-13 12:00:58 +00:00
Vladislav Vinogradov
442cd75c32 fixed compile error under linux 2010-12-13 09:50:32 +00:00
Vladislav Vinogradov
49ec8ba742 fixed bug in gpu filter engine (incorrect buffer type) and in vector's saturate_cast.
changed buffer type in linear filters to float.
added support of 1 channel image to linear filters.
added support of BORDER_REFLECT101, BORDER_REPLICATE and BORDER_CONSTANT border type to gpu linear filters.
minor fix in tests.
update comments in gpu.hpp.
2010-12-13 08:43:04 +00:00
Anatoly Baksheev
6c2cce6e28 some files rename 2010-12-11 15:21:11 +00:00
Anatoly Baksheev
ec4d2b6bb9 Fixed serious kernel crash IN StereoBM_GPU for some parameter. Will more rework for the code farther. 2010-12-11 15:07:14 +00:00
Alexey Spizhevoy
79b1eec3d3 minor refactoring of gpu module 2010-12-10 14:02:41 +00:00
Alexey Spizhevoy
9f80317ffa fixed minor bug in gpu module, added first version of sum 2010-12-10 13:36:00 +00:00
Alexey Spizhevoy
d557c800a7 refactored gpu module 2010-12-10 10:23:32 +00:00
Alexey Spizhevoy
97484089c5 added support of CV_8U & CV_TM_CCORR into gpu::matchTemplate 2010-12-10 08:06:54 +00:00
Alexey Spizhevoy
9801d07a46 added test for gpu:columnSum 2010-12-08 16:51:12 +00:00
Alexey Spizhevoy
fa322bf46f added gpu::columnSum, fixed compile error (if there is no cuda), refactored 2010-12-08 15:06:10 +00:00
Alexey Spizhevoy
b1c5b9293e added support of CV_32F & CV_TM_SQDIFF into gpu::matchTemplate 2010-12-08 13:12:12 +00:00
Alexey Spizhevoy
40304721a7 added support of CV_TM_CCORR (via FFT) into gpu::matchTemplate (versions both with block and without blocks) 2010-12-08 13:03:53 +00:00
Vladislav Vinogradov
905e5f1739 added support of 4-channels images to StereoConstantSpaceBP.
refactored transpose_gpu, made it non template function.
2010-12-08 07:23:59 +00:00
Vladislav Vinogradov
c18aa438ec added support of 4 channels images to StereoBeliefPropagation, minor code refactoring. 2010-12-07 10:20:03 +00:00
Alexey Spizhevoy
349a9fac86 fixed compile-time warning under g++ 2010-12-07 07:16:13 +00:00
Alexey Spizhevoy
19b434be50 minor refactoring of gpu module 2010-12-07 06:45:54 +00:00
Anatoly Baksheev
652fb1212e module reorganization: added folder with pure device functions, cuda_shared.hpp renamed to internal_shared.hpp 2010-12-06 16:37:32 +00:00
Vladislav Vinogradov
fadd19b976 added assertion on Compute Capability >= 1.1 to BruteForceMatcher_GPU_base::radiusMatch 2010-12-06 14:51:47 +00:00
Alexey Spizhevoy
50167f6c26 added first version of gpu::matchTemplate, currently it works only with 8UC1 images and SQDIFF method 2010-12-06 14:19:41 +00:00
Vladislav Vinogradov
8891acb67a added BruteForceMatcher_GPU 2010-12-06 12:06:51 +00:00
Alexey Spizhevoy
77027f6075 refactored border interpolation in gpu module 2010-12-06 09:44:51 +00:00
Vladislav Vinogradov
17d9014373 added per-element min/max to gpu module.
fixed compile error in transform.
2010-12-06 08:10:11 +00:00
Anatoly Baksheev
d96c5ebb7d misprint 2010-12-06 08:03:33 +00:00
Alexey Spizhevoy
6ef4d9b1dd added sipport of BORDER_REPLICATE mode into gpu::corner* functions 2010-12-06 07:47:26 +00:00
Alexey Spizhevoy
103bbaf09c updated BrdReflect101 in gpu module 2010-12-03 13:28:49 +00:00
Alexey Spizhevoy
e75ca4b662 replaced global memory reads with texture memory reads in GPU's corner detectors 2010-12-03 13:11:14 +00:00
Alexey Spizhevoy
df48ad76b0 added support of gamama_correction parameter into gpu::HOGDescriptor, updated tests 2010-12-03 11:11:44 +00:00
Alexey Spizhevoy
2a2590bae2 replaced filter (from the nearest to linear) mode when resizing image in gpu::HOGDescriptor 2010-12-03 10:14:01 +00:00
Alexey Spizhevoy
68c41a6c6c minor fix in gpu's header 2010-12-02 09:29:49 +00:00
Alexey Spizhevoy
ab04a03621 added support of BORDER_REFLECT into gpu::cornerHarris and etc, added support of border extr. int linear filters 2010-12-02 09:07:13 +00:00
Alexey Spizhevoy
d9489bbe0c fixed bug in gpu::cornerHarris and gpu::cornerMinEigenVal 2010-12-01 10:00:47 +00:00
Vladislav Vinogradov
57f7678db1 added gpu transpose for CV_8UC4, CV_8SC4, CV_16SC2, CV_16UC2, CV_32SC1 and CV_32FC1 types 2010-12-01 07:00:50 +00:00
Alexey Spizhevoy
d0dfae5143 fixed bug in ms segmentation 2010-11-30 13:02:24 +00:00
Alexey Spizhevoy
32a9b63d2f added better threads configuration estimator for the minMax, minMaxLoc, countNonZero functions in gpu module 2010-11-30 12:27:21 +00:00
Alexey Spizhevoy
493d7a6e45 added min eigen val based corner detector into gpu module 2010-11-30 08:44:04 +00:00
Alexey Spizhevoy
9adfc2cadc added Harris corner detector into gpu module 2010-11-30 08:04:37 +00:00
Alexey Spizhevoy
b260382840 added masks support into gpu::minMaxLoc 2010-11-29 13:56:43 +00:00
Alexey Spizhevoy
678f392569 added is_signed into numeric_limits_gpu, fixed incorrect min max finding for floating values 2010-11-29 13:21:43 +00:00