特征匹配与光流-浙江大学.pdf

特征匹配与光流 章国锋 浙江大学CADCG实验室 How can we find corresponding points? 稀疏对应——特征点匹配 根据匹配程度分为： 稠密对应——光流估计 What is a feature? Point Line: straight line, curve, Edge: 2D, 3D Corner Shape: rectangle, circle, ellipse, sphere, Texture Object Motion Invariance View point (scale, orientation, translation) Lighting condition Object deformations Partial occlusion What is feature extraction? From image space to feature space: Dimensionality reduction Find a latent semantic space where the pattern can be recognized easily Feature detection and representation Motivation For object recognition Motivation For object recognition Motivation Panorama creation Motivation Stereo pair matching Motivation Template matching Pick a template-rectangular/square region of an image Goal: find it in the same image/images of the same from different viewpoint Find corresponding points-Not always easy NASA Mars Rover images from “/Book/” Answer below (look for tiny colored squares) NASA Mars Rover images with SIFT feature matches Figure by Noah Snavely Interest points original Suppose you have to click on some point, go away and come back after I deform the image, and click on the same points again. Which points would you choose? deformed Corners • We should easily recognize the point by looking through a small window • Shifting a window in any direction should