2020-08-27 08:07:33 -0600 | received badge | ● Popular Question (source) |
2017-01-15 20:46:12 -0600 | received badge | ● Nice Question (source) |
2016-08-02 17:00:52 -0600 | received badge | ● Scholar (source) |
2016-07-31 14:08:56 -0600 | commented answer | stereo calibration issues Oh, nice catch, I did not even notice there was a stereoCalibrate call with both cameras calibration matrix... It gives a little improvement! I'm new to OpenCV and "just" need it to complete an obstacle map of the room. (I will also add an handmade sonar using a compass + IR range sensor) |
2016-07-31 10:47:51 -0600 | commented answer | stereo calibration issues Yes of course : http://warcraft3campagne.free.fr/Ster... I compute calibration of both cameras, then I undistort the images I use for the stereo calibration and pass them through stereoCalibrate |
2016-07-31 08:42:28 -0600 | received badge | ● Supporter (source) |
2016-07-31 08:35:16 -0600 | commented answer | stereo calibration issues This is a great answer too :) I was suspecting that my setup was not good enough to detect close items, I rebuilt my robot to have only 7cm between both cameras (instead of 14) and it seems MUCH better. (I will edit my post with results) After reading some stuff on internet ( here : https://www.cs.cf.ac.uk/Dave/Vision_l... ), it looks that it was a good thing to do. I tried the CV_CALIB_FIX_INTRINSIC but the RMS jumped to >2000. Right now, I'm keeping CV_CALIB_SAME_FOCAL_LENGTH and CV_CALIB_ZERO_TANGENT_DIST, the RMS is down to 0.28! |
2016-07-31 07:46:36 -0600 | received badge | ● Self-Learner (source) |
2016-07-31 06:05:15 -0600 | commented question | stereo calibration issues Hi, I converted a C++ sample in Java : http://warcraft3campagne.free.fr/Cali... Then I took 100 pics with a python script detecting chessboards. |
2016-07-31 05:24:44 -0600 | received badge | ● Student (source) |
2016-07-31 03:21:00 -0600 | received badge | ● Enthusiast |
2016-07-30 18:54:35 -0600 | asked a question | stereo calibration issues Hi! I'm currently working on a little robot with a stereo setup (made with 2 logitech C270, not the best, but still decent with 640x480 on linux) to detect obstacles indoor and create a 2D maps of the room. I tried to calibrate both cameras then calibrate the whole setup and I achieved finally to get decent results : Left camera with 100 images : Right camera with 100 images : Stereo calibration : RMS=0.3906354057793356 My number seems decent, even if the stereo calibration could be a bit better. I'm using those options for the camera calibration : And those ones for the stereo : So, my issue is that I can't get a decent disparity map. Even if the image seems to be undistored well : Original pic :
After undistortion + adding line to show Y axis :
But when it goes through the StereoBM.compute(), I can't get a good disparity map, even by tweaking the different values available : This is an example with those parameters : So, I would like to know if something looks wrong in my results, matrix, and datas, any help is welcome! Thank you EDIT : I made more tests this morning on a more opened area, and I'm getting better results : It looks like I'm able to get mid to long range datas (~4/5m and more) but It is struggling with low range. How can I improve this? Is it a calibration issue? The distance between both cameras is currently of 14cm. Is it too much? My goal would be to let the robot navigating among the room, so mid/long range detection is not really important, but close detection is! EDIT2: After reading internet stuff ( https://www.cs.cf.ac.uk/Dave/Vision_l... ), I understood that my setup (14cm between both cameras) was not good to find close items. I rebuilt my robot so I only have 7cm between both cameras and it looks much better! My first calibration gave me a RMS of 0.42, down to 0.28 ... (more) |
2015-08-16 06:37:20 -0600 | asked a question | stereo camera and depth map Hi guys, I want to create a little robot with a stereo camera (2x logitech c270). I made single calibration on both cameras (obtaining an error of 0.40 / 0.60 with 40 images each) then a stereo calibration (error of 0.5). What I don't undestand is that in different tutorials, it is said that using the My StereoBM parameters are : Am I missing something or there is some more work to do to get depth map from disparity mat? Thanks! |
2015-08-14 07:03:39 -0600 | commented question | initUndistortRectifyMap assert error Ah, nice to know.
I also found that there is a |
2015-08-13 19:45:57 -0600 | answered a question | How to pass a ArrayList<MatOfPoint> for a jni C ++ function? Hi, You should take a look to Calib3d JNI interface. For example the findFundamentalMat() function : And it's native part : So, this should do the trick : |
2015-08-13 19:16:12 -0600 | asked a question | initUndistortRectifyMap assert error Hi, I'm trying to rewrite the stereo calibrate example in Java. I think I did things okayish but I got an error I don't understand how it can happen : The correspond code : The error is coming from the initUndistortRectifyMap line, and from the d1 / d2 which are not empty and total() = 5 instead of 4. How this can happen whereas I didn't modify those Mat and I'm just using the returned Mat from stereoRectify() ? Any help would be welcome! Edit : If this can help, I'm using those flags to calibrate / findChessboard : |
2015-08-13 18:19:01 -0600 | commented question | imread returns empty mat in examples Ok, I found the issue, the configuration took my /usr/lib/libjpeg.so while the linker took another file. Adding "-DBUILD_JPEG=ON -DBUILD_TIFF=ON" to the cmake command made the trick. Hope it will help other people if they are facing the same problem. |
2015-08-12 18:06:14 -0600 | received badge | ● Editor (source) |
2015-08-12 16:46:33 -0600 | commented question | imread returns empty mat in examples It was just a check to see if the path was good. Of course it's not needed for the normal sample. |
2015-08-12 16:29:22 -0600 | asked a question | imread returns empty mat in examples Hi, I'm working on a little robot with stereo camera which communicates with a server (to compute depth map / find obstacles) I want to try the stereo calibration example of OpenCV 3.0. I had some issues to compile it but it looks ok now, ./configure is printing this output : (more) |