Vladislav Vinogradov
|
924670d32c
|
fixed block size calculation in SURF_GPU (fasthessian_gpu and nonmaxonly_gpu kernels)
|
14 years ago |
Anatoly Baksheev
|
7539b7de65
|
fixed build under ubuntu, but FS is still disabled
|
14 years ago |
Vladislav Vinogradov
|
5cd06d6a36
|
fixed SURF_GPU (fails on empty data)
added test for SURF_GPU and reprojectImageTo3D
|
14 years ago |
Anatoly Baksheev
|
f42a449df9
|
fix compilation under win32 with gpu
|
14 years ago |
Anatoly Baksheev
|
7b1c265563
|
fixed some compilation under ubuntu
|
14 years ago |
Anatoly Baksheev
|
21f0d1e174
|
removed MS specific from FD code
|
14 years ago |
Anatoly Baksheev
|
0747f2d863
|
1) NPP_staging as sources. Binaries removed.
2) NVidia tests for GPU
3) FD sample that uses NVidia's interface.
|
14 years ago |
Alexey Spizhevoy
|
50429d8a3e
|
fixed some warnings and errors under g++
|
14 years ago |
Alexey Spizhevoy
|
ea94b43541
|
added stereo_multi_gpu sample, cosmetic changes in multi_gpu sample
|
14 years ago |
Alexey Spizhevoy
|
557dd39f03
|
fixed gpu::sum* on CC1.0, updated some tests
|
14 years ago |
Alexey Spizhevoy
|
e303b0dd62
|
added CUDA_ARCH_FEATUERS define (for correct handling of BIN(PTX) cases)
|
14 years ago |
Alexey Spizhevoy
|
16e74ab306
|
added buffered version of norm, updated performance sample and docs
|
14 years ago |
Alexey Spizhevoy
|
3795142604
|
implemented gpu::norm via absSum, sqrSum, and minMax (removed norm_diff call), added support of other data types
|
14 years ago |
Alexey Spizhevoy
|
ae529f4bc6
|
added absSum function
|
14 years ago |
Alexey Spizhevoy
|
fa446e7e35
|
removed linear_filters_beta.cu as its functionality was moved into filters.cu
|
14 years ago |
Vladislav Vinogradov
|
8274ed22e4
|
fixed gpu tests (BruteForceMatcher_GPU, divide, phase, cartToPolar, async)
minor code refactoring
|
14 years ago |
Alexey Spizhevoy
|
7a29d96cf4
|
added buffered version of gpu::integral function and updated performance test (it still works too slow)
|
14 years ago |
Alexey Spizhevoy
|
0af5356dbc
|
updated code in case of the compilation without CUDA
|
14 years ago |
Alexey Spizhevoy
|
575fd1fe4c
|
reafactoring: replaced query device props functions with the DeviceInfo class
|
14 years ago |
Alexey Spizhevoy
|
65b9f3bc10
|
fixed TargetArchs implementation in case when HAVE_CUDA=false, added initial structure for multi_gpu sample
|
14 years ago |
Alexey Spizhevoy
|
891e2ff310
|
replaced has* methods in the GPU module with the TargetArchs monostate
|
14 years ago |
Alexey Spizhevoy
|
13a6d0b92a
|
fixed parsing GPU archs in BIN(PTX) format
|
14 years ago |
Vladislav Vinogradov
|
eda8416358
|
fixed BruteForceMatcher_GPU (fails if input data is empty)
updated BruteForceMatcher_GPU test
|
14 years ago |
Alexey Spizhevoy
|
310c483da8
|
added first version of gpu performance tests
|
14 years ago |
Vladislav Vinogradov
|
811ba31897
|
added synchronization after NPP calls
|
14 years ago |
Vladislav Vinogradov
|
8abdb3721f
|
added gpu threshold.
|
14 years ago |
Alexey Spizhevoy
|
055c226392
|
fixed FindNPP.cmake Apple-aimed part
|
14 years ago |
Alexey Spizhevoy
|
e51b9021bc
|
forgot to rename namespace
|
14 years ago |
Alexey Spizhevoy
|
db41449be8
|
fixed errors under MacOS
|
14 years ago |
Anatoly Baksheev
|
e3f3de84db
|
warning fixed
|
14 years ago |
Alexey Spizhevoy
|
01dafce1a1
|
fixed some bugs in GPU matrix reductions, removed <functional> into precomp.hpp
|
14 years ago |
Alexey Spizhevoy
|
0da71a01ff
|
fixed some GPU tests failing when compiled for 1.1(no doubles) and run on 1.3(with doubles)
|
14 years ago |
Alexey Spizhevoy
|
9e48f64149
|
fixed parsing of CC in gpu module
|
14 years ago |
Alexey Spizhevoy
|
574b3f94a1
|
updated gpu initialization functions, added compile-time error on CC 1.0
|
14 years ago |
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
|
14 years ago |
Alexey Spizhevoy
|
4b8425dbb8
|
fixed minor bugs
|
14 years ago |
Alexey Spizhevoy
|
186b1fc6ef
|
refactored gpu module
|
14 years ago |
Alexey Spizhevoy
|
90ae1e3aed
|
refactored gpu module
|
14 years ago |
Alexey Spizhevoy
|
1a0d41fb53
|
added checkPtxVersion into gpu module
|
14 years ago |
Alexey Spizhevoy
|
cbb132ccb1
|
added ensureSizeIsEnough into gpu module, updated reduction methods
|
14 years ago |
Alexey Spizhevoy
|
f3a2656808
|
added hasPtxFor and isCompatibleWith functions into gpu module, added docs for them
|
14 years ago |
Anatoly Baksheev
|
cb63046dcf
|
gpu face detetcion:
1) fixed bug with error codes (enum NppStStatus) shift.
2) added some asserts
|
14 years ago |
Alexey Spizhevoy
|
db852e0b54
|
updated gpu module docs
|
14 years ago |
Vladislav Vinogradov
|
349e0ece93
|
added docs for GPU Filter Engine
|
14 years ago |
Anatoly Baksheev
|
1a94186195
|
First version of CascadeClassifier_GPU.
Only for VS2008 now.
Sample for it.
new NPP_staging for VS2008 only
|
14 years ago |
Alexey Spizhevoy
|
4e23f37ff8
|
added docs for gpu::HOGDescriptor
|
14 years ago |
Alexey Spizhevoy
|
4dfbf99dd5
|
cosmetic changes in gpu module, decreased matchTemplate test running time
|
14 years ago |
Alexey Spizhevoy
|
d091ae5746
|
removed unnecessary code from gpu::matchTemplate
|
14 years ago |
Alexey Spizhevoy
|
57f917d6f2
|
replaced adding constant in normalization with taking max
|
14 years ago |
Alexey Spizhevoy
|
dc763e0250
|
updated normalization routine in the matchTemplate to avoid division by zero on black images (ticket #798), added test
|
14 years ago |