Harris corner detector and corner stability
WHAT IT TESTS: feature detection intuition. OUTLINE: Harris finds points where intensity changes strongly in all directions using the structure tensor of gradients; corners are well localized in two directions, unlike edges.
WHAT IT TESTS: whether you understand why corners are good trackable features. ANSWER OUTLINE: Harris examines a window and computes how much the intensity changes when the window shifts, summarized by the structure tensor built from image gradients; a corner produces large change in every direction (both eigenvalues large), an edge changes in only one direction, and flat regions change in none. Corners are localizable in two dimensions, avoiding the aperture problem that makes edges slide. RED FLAG: equating corners with edges.
Read the original → interview
- #harris-corner
- #feature-detection
- #structure-tensor
- #tracking
- #aperture-problem
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.