2015-04-16 18:52:45 -0600 | asked a question | Opencv with IPP remap function support Hi, Does OpenCV with IPP supports remap function(speed up remap function)? Best regards Viktor. |
2015-04-06 23:00:20 -0600 | commented question | transpose matrix for remap Thanks. I will ask emgu support. |
2015-04-06 14:26:34 -0600 | commented question | transpose matrix for remap I am using emgu. Emgu hasn't CudaInvoke.Transpose. Now i am doing: mapx[i,j]=j; mapy[i,j]=i; CvInvoke.Remap is ok, but CudaInvoke.Remap return black image It is bug? |
2015-04-06 12:48:21 -0600 | asked a question | transpose matrix for remap Hi, I want remap image like transpose function. Anybody knows matrix or algorithm for transposing source matrix? Best regards Viktor. |
2015-02-12 04:26:10 -0600 | answered a question | opencv 3.0 stitcher error on CUDA If i use FeaturesMatcher without CUDA, then no error: Ptr<detail::featuresmatcher> featuresMatcher = makePtr<detail::bestof2nearestmatcher>(false, 0.65f); stitcher.setFeaturesMatcher(featuresMatcher); But stitch with CUDA not faster than without CUDA in opencv 3.0 beta. Also default blender/seamfinder with CUDA not good. |
2015-02-10 00:12:30 -0600 | received badge | ● Enthusiast |
2015-02-09 07:06:59 -0600 | asked a question | opencv 3.0 stitcher error on CUDA Hi, stitcher = Stitcher::createDefault(true); Mat pano; vector<mat> imgs; imgs.push_back(img1); imgs.push_back(img2); stitcher.stitch(imgs, pano); Last line throws error: OpenCV Error: Assertion failed (func != 0) in knnMatchSingle, file /home/hyper/Downloads/opencv-3.0.0-beta/modules/cudafeatures2d/src/brute_force_matcher.cpp, line 503 terminate called after throwing an instance of 'cv::Exception' what(): /home/hyper/Downloads/opencv-3.0.0-beta/modules/cudafeatures2d/src/brute_force_matcher.cpp:503: error: (-215) func != 0 in function knnMatchSingle Stitching without GPU - OK. Any ideas? Thank you. |
2015-02-09 02:34:42 -0600 | received badge | ● Self-Learner (source) |
2015-02-07 01:57:11 -0600 | received badge | ● Editor (source) |
2015-02-07 01:34:10 -0600 | answered a question | Opencv 3 compiled with CUDA but GetCudaEnabledDeviceCount always 0 Resolved. Opencv 2.4.9 same result. But if running under root DEVICE DETECTED Also gcc4.8 not playing with CUDA together. I use gcc4.7 Thank you all. CLOSED. |
2015-02-03 22:06:45 -0600 | asked a question | Opencv 3 compiled with CUDA but GetCudaEnabledDeviceCount always 0 Hi, I want use CUDA acceleration for my opencv based project. I have installed latest nvidia drivers, nvidia toolkit, opencv 3 beta with_cuda=on But getCudaEnabledDeviceCount is always return 0 Please help Nvidia deviceQuery is here: http://verti.ee/etc/deviceQuery.txt Compiler output is here: http://verti.ee/etc/make.txt Best regards Viktor |