A DRAFT computer vision book by Prof. Richard Szeliski. The book reflects the author's wide experience in practical computer vision algorithms that he has developed while working in corporate research labs, mostly at Digital Equipment Corporation's Cambridge Research Lab and at Microsoft Research. The book focuses more on practical basic techniques that work under real-world conditions, and hence is a suitable book for researchers working in industry. Note: This is a draft book. Most of the chapters are not complete yet. The book covers the following topics: Image formation, Image processing, Feature detection and matching, Geometric alignment and calibration, Structure from motion, Dense motion estimation, Image stitching, Computational photography, Stereo correspondence, Shape and appearance modeling, Image-based rendering, Video-based rendering, and Recognition.