Detectors and Descriptors

This chapter contains recipes for:

• Finding corners in an image - Harris and FAST
• Selecting good corners in an image for tracking
• Drawing keypoints, descriptors, and matches
• Detecting scale invariant keypoints
• Computing descriptors for image keypoints - SURF, BRIEF, and ORB
• Matching techniques for finding correspondences between descriptors
• Finding reliable matches - cross-check and ratio test
• Model-based filtering of matches - RANSAC
• BoW model for constructing global image descriptors