I want to visualize keypoints on images. However, keypoints obtained from KPDetector doesn't make sense to me as they are values around zero. What are the scaling and translation values that I should use for scaling them to correct spatial positions?
I want to visualize keypoints on images. However, keypoints obtained from KPDetector doesn't make sense to me as they are values around zero. What are the scaling and translation values that I should use for scaling them to correct spatial positions?