Computer Vision News - March 2022

16 Computer Vision Book Richard, how did you come to write the book? In 2003, Steve Seitz invitedme to co-teach a class at theUniversity ofWashington, which we called ‘Computer Vision for Computer Graphics’. Several techniques based on computer vision and analyzing images could be used to create computer graphics effects or 3D models. In 2005, we taught a more general computer vision class, and I decided to convert all that knowledge we were sharing into a book. It finally came out in 2010. That first edition followed the traditional layering of low-level, mid- level, and high-level vision in those days, starting with basic things like projective geometry, then image processing and 2D and 3D vision, and finally recognition. That whole idea has been upended with deep learning. We still have different layers, but Richard Szeliski is an Affiliate Professor at the University of Washington, where he supervises students and teaches some lectures. He retired from Facebook just over a year ago, after a career spanning 40 years in computer vision research, but that has not stopped him. He speaks to us after publishing the second edition of his all-encompassing computer vision textbook, ‘Computer Vision: Algorithms and Applications’. COMPUTER VISION: ALGORITHMS AND APPLICATIONS