Project Description
This is a forked repository. The original OpenCV repository is here.
The PR I'm Currently Working On
[#22041] remove const from functions returning by value
List of Merged PRs (opencv)
[#19259] Minimal implementation of GpuMatND (test code in contrib)
[#18136] bit-exact cuda::equalizeHist
[#17982] cuda::GpuMat::convertTo - fix for in-place arguments
[#17863] cuda::flip - Use in-place npp function for inplace arguments (port to contrib)
[#17556] cuda optflow TVL1 : run safely in async mode
[#14041] Extract Ptr-related code from lut.cu to new lut.cpp
[#14000] CUDA 10.1 Build Issue Fix on Master Branch
[#13960] Windows Build Issue Fix
(20)
[#13958] CUDA 10.1 Build Issue Fix
[#13903] cudev - Rework some code
[#13850] cuda::StereoBM - Fix hanging and racing issue
[#13810] cudalegacy: Use safe block scan function
[#13764] Add CV_16UC1 support for cuda::CLAHE
[#13658] __shfl_up_sync with mask for CUDA >= 9
[#13625] Fix Farneback Optical Flow Algorithm
[#13596] Remove build warning msg with CUDA 10.0
[#13364] Fix error in LineIterator example code in doc
[#11572] NPP : NppStreamHandler fix
(10)
[#11526] cuda_meanStdDev : bug fix
[#11483] cuda_canny : multi stream safety
[#11155] Update GpuMat, GpuMat::download, GpuMat::upload documentation
[#11110] test_buffer_pool: synchronize after async copy
[#10987] SSE2 : use _mm_cvtpd_epi32 when converting from CV_64F to CV_32S
[#10906] cuda_fast : multi stream safety
[#10861] Fix for CUDA_Arithm/Dft.Algorithm/0 test
[#10803] Update BufferReader documentation with some example code
[#10751] cuda_stream: do not allocate GPU memory by default
[#10640] cv::cuda::cvtColor bug fix
List of Involved PRs (opencv)
[#19286] (move semantics)add cuda::Stream constructor from cudaStream_t
[#17432] CUDA: choose supported CC automatically
[#12722] cudafilters: fix test failure of cudafilters Median_Accuracy
[#11064] cudaarithm: make the asynchronous call to NPP safe
List of Involved Discussions (opencv, opencv_contrib)
[#18347] cudaarithm: inplace version of NPP flip fails with odd number ROI
(30)
[#18155] cuda_OpticalFlowDual_TVL1 is not thread-safe in python
[#18051] CUDA GoodFeaturesToTrackDetector is not ThreadSafe ?
[#18035] CUDA equalizeHist does not produce identical result
[#17840] In-place flip of GpuMat produces image artifacs
[#13092] cv::cuda::GpuMat.convertTo() seems not to support in-place, while cv::Mat does
[#16013] Corrupted optical flow using cuda::DenseOpticalFlow asynchronously in multithreaded environment
[#2361] (contrib)Bug in cv::cuda::warpPerspective
[#14052] an illegal memory access was encountered in function 'download'
[#14017] Opencv 4.0.1 with Cuda
[#13996] opencv-4.0.1 ,CUDA10.1, failed to build cudaimageproc
(20)
[#13984] Problem compiling clahe.cu - identifier "PtrStepus" is undefined
[#13952] OpenCV 4.0.1 + Cuda 10.1, failed to build?
[#13897] Failed to build OpenCV 4.0.1 with CUDA 10 10.0
[#13883] Template Matching is not threadsafe
[#13761] cudalegacy NCVHaarObjectDetection hangs with RTX 2080 Ti
[#13491] Error when building with CUDA. VS 2017, Win10.
[#1958] (contrib)Feature request: Cuda CLAHE for 16 bit images
[#13477] cuda::createTemplateMatching not work with CUDA10.0
[#13014] cuda blockScanInclusive hangs with RTX 2080
[#12895] cudaoptflow: test failure of FarnebackOpticalFlow
(10)
[#12721] cudafilters: Median_Accuracy fails with CUDA 9.0 and after
[#12320] cv::cuda::integral hangs on Titan V
[#11622] CUDA Median filter tests fail with CUDA 9.1 but pass with CUDA 8.0
[#11606] Suggestion for the CUDA stream module
[#11511] unneeded cudaStreamSynchronize(stream_)
[#11298] bug in MemoryReturn in cuda module
[#11063] cudaarithm: async call to NPP fails
[#8938] Can –default-stream per-thread be used with opencv ?
[#8725] Calling cv::cuda::Stream::Null() results in a stray cudaMalloc() call
[#6742] cv::cuda::Filter thread safety
Waiting for Response - For my own reference :)
[#11606] Suggestion for the CUDA stream module
[#12585] cuda: move CUDA modules to opencv_contrib