third_party/opencv
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
a75840d19c
opencv/modules
History
Arjan van de Ven a75840d19c Merge pull request #10468 from fenrus75:avx512-2 
* Add a 512 bit codepath to the AVX512 fastConv function

this patch adds a 512 wide codepath to the fastConv() function for
AVX512 use.
The basic idea is to process the first N * 16 elements of the vector
with avx512, and then run the rest of the vector using the traditional
AVX2 codepath.

* dnn: use unaligned AVX512 load (OpenCV aligns data on 32-byte boundary)

* dnn: change "vecsize" condition for AVX512

* dnn: fix indentation
2018-01-31 16:34:12 +03:00
..
calib3d Merge pull request #10667 from paroj:stereo_calib_ex 2018-01-26 22:05:13 +03:00
core core:ppc Add yield support 2018-01-31 04:03:35 +00:00
cudaarithm Fix perf build with CUDA 9. 2018-01-19 00:23:02 +01:00
cudabgsegm
cudacodec
cudafeatures2d
cudafilters
cudaimgproc Fix perf build with CUDA 9. 2018-01-19 00:23:02 +01:00
cudalegacy clean up the code 2017-12-22 08:42:21 +09:00
cudaobjdetect
cudaoptflow
cudastereo Fix perf build with CUDA 9. 2018-01-19 00:23:02 +01:00
cudawarping Fix perf build with CUDA 9. 2018-01-19 00:23:02 +01:00
cudev cv::cuda::cvtColor bug fix (#10640) 2018-01-19 14:06:05 +03:00
dnn Merge pull request #10468 from fenrus75:avx512-2 2018-01-31 16:34:12 +03:00
features2d fix build error on Visual Studio 2013 and earlier 2018-01-31 18:22:09 +09:00
flann resolves #10548 - FLANN::knnSearch garbage bug (when kNN is larger than the dataset size) 2018-01-08 18:58:55 +09:00
highgui Merge pull request #10649 from GregoryMorse:patch-3 2018-01-21 09:26:48 +03:00
imgcodecs Merge pull request #10512 from sturkmen72:update_documentation 2018-01-18 04:44:59 +00:00
imgproc solve issue #10687 2018-01-26 17:01:02 +01:00
java Merge pull request #10621 from mshabunin:disable-docs 2018-01-18 16:37:19 +03:00
js JavaScript bindings for dnn module 2017-12-08 18:33:48 +03:00
ml ml(ANN_MLP): ensure that train() call is always successful 2017-12-22 18:50:04 +00:00
objdetect Implementation of bit-exact resize. Internal calls to linear resize updated to use bit-exact version. (#9468) 2017-12-13 15:00:38 +03:00
photo Reverted calls to linear resize back to generic version for floating point matrices 2017-12-22 17:10:03 +03:00
python persistence: replace arbitrary limit of cn to 4 by CV_CN_MAX (#10636) 2018-01-23 13:36:30 +03:00
shape Implementation of bit-exact resize. Internal calls to linear resize updated to use bit-exact version. (#9468) 2017-12-13 15:00:38 +03:00
stitching Exported a high level stitcher the DLL 2018-01-24 12:43:00 -08:00
superres
ts Merge pull request #10691 from alalek:parallel_for_2018 2018-01-30 14:13:29 +00:00
video ocl: Avoid unnecessary initializing when non-UMat parameters are used 2018-01-21 23:02:05 +08:00
videoio java(test): fix test names 2018-01-12 19:14:00 +00:00
videostab Merge pull request #9979 from fakabbir:ImproveDoc 2017-11-08 16:55:48 +00:00
viz handle legacy VTK functions 2017-12-21 01:33:25 -08:00
world
CMakeLists.txt Merge pull request #10621 from mshabunin:disable-docs 2018-01-18 16:37:19 +03:00