x10s's profile - activity

That helped, thanks! They had a very good depiction of the two valid conditions and how they work.

Delta is usually used in math to refer to a change in value e.g x +/- delta

A simple search in the documentation gave me this http://docs.opencv.org/2.4/modules/im...

scale is the scaling factor for the computed derivatives while delta is the delta that can be added to the output image before it is stored. borderType should be the way the border pixels are extrapolated(I'm not a 100% sure on this)

dll? Your original question says that you're on a linux machine and your file paths indicate the same. Have you followed the correct compile and install procedure for opencv on linux?

No, that's what I meant. The only other thing I can think of are unmet dependencies like ffmpeg?

I realized that my purely rotating camera has some translation which maybe the cause for the Homography to be bad overall. My scene does not have a planar surface as of now. I shall be testing my hypothesis today. This link states the two conditions when a Homography is valid but doesn't explain it very well :(

Does your program have read permissions on the video file?

I missed this but I hope you skipped checking out the release branch for opencv_contrib as well. They need to be the same version.

Clone the opencv repo to a new folder and don't do the checkout step i.e.

git checkout 3.0.0

The rest of the steps can be followed as mentioned in the link. This means you will be working with the dev branch instead of the release branch. The dev branch has the fix.

Have you tried using the matched image co-ordinates from the two images of the planar surface and computing the homography to calculate the euler angles(i.e. pitch, yaw and roll) ?

P.S Apologies for the throwback as I just noticed this question is from more than a year ago.

You should check if the video you're trying to read is opened before trying to read a frame and convert it to grayscale.

  cap = cv2.VideoCapture("/home/user/.VIRTUALENVS/cv/lib/python2.7/site-packages/ped1.avi")
  while(cap.isOpened()):
    ret, frame1 = cap.read()
    if ret:
      prvs = cv2.cvtColor(frame1,cv2.COLOR_BGR2GRAY)

Thank you for your input. I shall edit my question to be more specific :)

Hello,

I am computing the Homography between two images and using that to estimate the change in camera pose.

1. The images are being taken by a camera rotating about the z-axis(or a purely rotating camera)
2. I use a feature detection method e.g. SIFT/SURF/ORB and use the Flann matcher to get the putative matches. On visual inspection the matches look correct.
3. The outlier detection method being used is RANSAC with a low reprojection threshold of 1.0
4. When the yaw between images taken is larger than 15 degrees or so, the matches look right on visual inspection but the average reprojection error((I calculate this myself using x' - H*x)) goes high. And unfortunately its not just a few outliers, the reprojection error is high with all the matches.
5. The matches are the close to the image boundaries i.e. in the first image the interest points are close to the right boundary and in the second image its close to the left boundary. Not sure if this changes anything though.

Are there any guidelines to improve the re-projection error(and so the Homography)? Or does anyone know of any literature that could help?

Thanks!