2019-08-13 06:48:50 -0600 | received badge | ● Notable Question (source) |
2018-02-11 21:49:33 -0600 | asked a question | Should OpenCVConfig.cmake be part of dev debian package? Should OpenCVConfig.cmake be part of dev debian package? Hi, I am building my own OpenCV debian packages and when I wan |
2016-06-21 19:20:18 -0600 | commented question | Real time pedestrian detection ok that's a good point but for the accuracy you should plot a DET curve otherwise it's hard to be sure which one is the best. Also you can have a look at this paper it's comparing different algorithm. http://www.vision.caltech.edu/publica... |
2016-06-20 20:45:15 -0600 | commented question | Real time pedestrian detection When you say your results are better than opencv can you give more explanations? Did you plot a DET curve? Have you tried on a lot of of test samples? OpenCV log-average miss rate gives me the same results as the INRIA HOG optimised by the authors (46%) |
2016-06-20 20:43:01 -0600 | commented answer | SVM bias on weights of positives and negatives No worries I actually answer to my question myself. Looks like HOG OpenCV can't get better than 46% log-average miss-rate. I achieved the same results with my own SVM. However it's not accurate enough. I found the paper comparing the different pedestrian detection algorithm so I will probably have a look at that! Any hint on some algorithm you tried and work better? |
2016-06-16 19:53:42 -0600 | received badge | ● Supporter (source) |
2016-06-13 18:59:59 -0600 | received badge | ● Enthusiast |
2016-06-09 22:33:55 -0600 | commented answer | SVM bias on weights of positives and negatives @StevenPuttemans Hey Steven, I'm wondering what kinda results you got at the end. I've computed some accuracy using DET curve with the INRIA dataset. The only difference are the detection parameters are scale and window. strides. Using the HOG INRIA parameters (1.2, 8x8) gives poor results on the INRIA dataset. Using HOG Opencv parameters (1.05, 4x4) gives better results. I'm not sure how they have manage to get good accuracy with scale that high. But anyway the best I could get is 0.35 (for FNR = FPPW). Both used the default opencv SVM |
2016-03-16 10:20:44 -0600 | received badge | ● Popular Question (source) |
2016-01-05 18:13:01 -0600 | commented question | Building static OpenCV with ffmpeg is still looking for shared libavcodec ok thanks. Is there any other library i can use other than ffmpeg thread video? and which would be under BSD?. I saw GStreamer is also LGPL |
2016-01-05 00:23:31 -0600 | received badge | ● Editor (source) |
2016-01-05 00:22:38 -0600 | asked a question | Building static OpenCV with ffmpeg is still looking for shared libavcodec I'm trying to build a static OpenCV python with ffmpeg enabled to run on AWS Lambda. I Need to upload OpenCV my cv2.so in a zip file. To do so I set the BUILD_SHARED_LIBS=OFF and WITH_FFPMEG=YES. But when doing import cv2 in the python shell, I get the error: ImportError: libavcodec.so.53: cannot open shared object file: No such file or directory. If I set WITH_FFPMEG=NO I don't get this error anymore but I need ffmpeg to read a video. What I would like to do is opencv to use the static libavcodec library not the shared one. I thought setting up BUILD_SHARE_LIBS to off would be enough but it's not. There are no flags in the cmake file which let you specified the path to libavcodec. I've been looking for sometimes but I can't find a solution. Any ideas? Thanks for your help |
2015-12-30 21:39:44 -0600 | commented question | How to build opencv so it statically link to ffmpeg? How can you tell opencv to use this static ffmpeg? there is no option in the cmake to link ffmpeg? |
2015-12-29 07:06:56 -0600 | asked a question | Building static openCV still uses shared libraries Hi, I'm trying to build a static opencv library to run on AWS Lambda. To do so I put the BUILD_SHARED_LIBS to OFF. However it is still using libgstreamer-0.10.so when linking and not libgstreamer-0.10.a. (There is no option to point at the static one with cmake). Any idea why? Both libraries are present on my machine and i'm using Ubuntu 11.0. Thanks for your help |
2014-06-26 02:00:41 -0600 | asked a question | BackgroundSubtractorMOG2 input Hi, I just would like to know if I can use an grayscale image as an input with BackgroundSubtractorMOG2 or only a RGB image. I couldn't find anything in the documentation or the paper saying it has to be a RGB image, but I would like to be sure. In what I saw with a sample video, the sum of the Foreground pixels of RGB = sum Foreground pixel of Gray + constant Thanks for your help! |
2014-04-07 22:18:12 -0600 | asked a question | Error building openCV 2.4.9 library not found for -lopencv_cudev Hi all, I'm having trouble to build opencv on my Mac OS 10.9.2 to us it with python During the make step I get the following error: and even if a disable BUILD_opencv_python And I'm also getting this. Not sure if it's related or not Any idea? Thanks for your help! (more) |