2020-04-15 04:55:21 -0600 | received badge | ● Enthusiast |
2020-04-11 08:02:01 -0600 | commented question | LSH knnSearch slow in large dataset looking at C++ code seems that distance type can be passed as 3rd parameter to build function of any Index. HAMMING shou |
2020-04-10 10:04:08 -0600 | received badge | ● Supporter (source) |
2020-04-10 10:03:54 -0600 | commented question | iOS OpenCV: Image matching with ORB leads to EXC_BAD_ACCESS I found the culprit. cv::ORB *orb = cv::ORB::create(); should be cv::Ptr<cv::ORB> orb = cv::ORB::create(); othe |
2020-04-10 03:17:41 -0600 | commented question | iOS OpenCV: Image matching with ORB leads to EXC_BAD_ACCESS I'm facing same issue, did you finally solve that? I'm use OpenCV 4.3 ... but I'm not sure if contrib is required for de |
2020-04-09 12:11:14 -0600 | commented question | LSH knnSearch slow in large dataset Thanks @berak but scikit is a lot slower (500x in KDTree implementation). I've also tried OpenCV Kmeans (using float des |
2020-04-09 12:10:18 -0600 | commented question | LSH knnSearch slow in large dataset Thanks @berak but scikit is a lot slower (500x in KDTree implementation). I've also tried OpenCV Kmeans (using float des |
2020-04-09 12:09:42 -0600 | commented question | LSH knnSearch slow in large dataset Thanks @berak but scikit seems to be a lot slower (500x in KDTree implementation). I've also tried OpenCV Kmeans (using |
2020-04-09 04:56:39 -0600 | commented question | LSH knnSearch slow in large dataset @hha1 each image should have more than 1 descriptor! In my case 10K images means a few million descriptors (how many dep |
2020-04-09 04:55:48 -0600 | commented question | LSH knnSearch slow in large dataset @hha1 each image should have more than 1 descriptor! In my case 10K images means a few million descriptors (how many dep |
2020-04-09 04:55:34 -0600 | commented question | LSH knnSearch slow in large dataset @hha1 each image should have more than 1 descriptor! In my case 10K images means a few million descriptors (how many dep |
2020-04-08 13:43:12 -0600 | commented question | LSH knnSearch slow in large dataset @hha1 look at cv2.flann_Index and knnSearch, I can post code if you are more specific about your problem |
2020-04-08 13:41:06 -0600 | commented question | LSH knnSearch slow in large dataset @berak yes I’m saving keypoints, descriptors and index. At startup I load them and then I start processing query images |
2020-04-08 10:44:44 -0600 | received badge | ● Student (source) |
2020-04-08 10:32:17 -0600 | asked a question | LSH knnSearch slow in large dataset LSH knnSerach slow in large dataset I'm using python version of OpenCV 4.2 and I'm trying to do feature matching inside |