Feature Extraction And Matching

Feature Extraction and Matching
Contact: Fernando Alonso-Fernandez

Welcome to this lecture on Feature Extraction and Matching Techniques.

In this lecture we will give an overview of algorithms for feature extraction and matching. The purpose of features is to extract particular properties (features) of images, intended to be non redundant and informative enough to enable subsequent learning tasks. The list learning tasks is huge, including (to cite just a few) person or object recognition, motion tracking, pose estimation, robot navigation, 3D reconstruction, etc.

Teaching Material

Get slides from Google docs (ppt) here

References and sources

R. Klette, “Concise Computer Vision”, Springer, 2014 Sections 9.1, (feature invariance), 9.2 (SIFT and others, testing invariance) Section 9.3 (tracking and updating of features), not seen here Get chapter 9 Site of the author (full book not available, but plenty of other resources): http://ccv.wordpress.fos.auckland.ac.nz/

R. Szeliski, “Computer Vision: Algorithms and Applications”, Springer 2010 Section 4.1 (features) Full book available online: http://szeliski.org/Book/