Features in Computer Vision - Lecture 14

Explore the fundamentals of features in computer vision in this lecture from the University of Central Florida's CAP5415 Computer Vision course. Delve into the definition and types of features, their applications in matching, structure from motion, and panorama stitching. Learn about finding features in videos and the characteristics of good features, including distinctiveness, compactness, and efficiency. Examine the process of choosing interest points, their properties, and approaches like corner detection. Discover the goals for keypoints and their practical applications in keypoint matching. This comprehensive lecture provides essential knowledge for understanding feature extraction and its role in various computer vision tasks.

Outline
What is a Feature?
Types of Features
Uses of Features: Matching
Uses of Features: structure from motion
Uses of Features: panorama stitching . Given two images . How do we overlay them?
Finding Features in Videos
Characteristics of good features Distinctiveness Each feature can be uniquely identified
Compactness and Efficiency
Choosing interest points
What is an interest point?
Properties of Interest Points
Possible approaches: corner detection
Goals for KeyPoints
Application: Key Point Matching
Application: KeyPoint Matching