snavas's profile - activity

Thanks for the suggestion, I will try that out!

I was making use of convexityDefects for hand detection/recognition in my project using opencv-contrib-python version cv2 4.2.0.32. I recently updated my cv2 version to the latest one, but my functionality broke.

error: (-5:Bad argument) The convex hull indices are not monotonous, which can be in the case when the input contour contains self-intersections in function 'cv::convexityDefects'

I was able to backtrack and find the version where convexityDefects stopped behaving as I was expecting (as it behaved before). In 4.2.0.34 it still works but from version 4.3.0.36 and on I get the hull indices are not monotonous error.

Do someone knows if the implementation of the convexityDefects function changed across versions and what did it change?

Aditional information:

• The contour that throws the error
• opencv-contrib-python 4.2.0.32/34 vs 4.3.0.36>
• Windows 10 10.0.18363 x64

Thanks!

So, as I understand, opencv did not check before if the convex hull indices were monotonous or not. However, the hull in

Did convexityDefects function behaviour changed between cv2 versions 4.2.0 and 4.3.0? I was making use of convexityDefec