Lucas-Kanade Method: Tracking Pixel Patches, Not Points

The Lucas-Kanade method tracks motion by assuming a small patch of pixels moves as a single rigid unit. Instead of asking "where did this one pixel go?", it asks "what single motion vector best explains the intensity changes in this whole neighborhood?". This is a cornerstone of computer vision for estimating optical flow, used in video stabilization and feature tracking for AR. The footgun: its core assumptions are its weakness. It fails with large displacements or significant lighting changes, as it relies on brightness constancy.