Instructor:Deva Ramanan (dramanan@ics.uci.edu)
Office: DBH 4072, 842-4893
Office hours: Tuesday 1:00-2:30pm or by appointment
Lectures: MWF 11:00-11:50pm ICS 180
Course description
This course focuses on the classic computer vision problems of light transport and multiview geometry. The driving focus is to undertand the three-dimensional content of an image by analyzing the illumination conditions and geometry of multiple cameras. The first half of the course will discuss the physics of light transport, and its interaction with 3D shape and image formation. The second half of the course deals with reconstruction from multiple camera views. This course will review much of the basic theory of illumination models and projective geometry. The material will be of relevance to students interested in vision and graphics, as well as those interested in specific applications of geometric image understanding and multi-camera networks.
Prerequisite:
ICS 6D/Mathematics 6D, Mathematics 6G or 3A, Mathematics 2A-B, ICS 23, ICS211A. Please see me if you haven't taken these classes.
Textbooks:
Lectures will be based on a variety of source, including textbooks and recent research papers. Only the first textbook is officially required, but all are recommended. All have been placed on reserve in the library
1. "Computer Vision: A Modern Approach" by Forsyth and Ponce.
2. "Multiple View Geometry" by Zisserman and Hartley.
3. "An Invitation to 3D Vision" by Ma et al.
4. "Robot Vision" by Horn.
Grading:
1. There will be 3-4 programming assignments worth a total of 40% of the final grade.
2. Each student will be assigned to scribe 1 to 2 lectures in latex. These are due 1 week after the assigned lecture. This will be worth a total of 10% of the final grade.
3. There will be a final project worth 50% of the final grade.
Collaboration policy / Academic honesty:
Homeworks can be discussed, but each student must independently write up their own solutions. In particular, no sharing of code. Please see the university policy on academic honesty.
It is fine to use reference materials found online, but do not search for homework solutions. Rather, students are strongly encouraged to ask questions at both office hours and on the class discussion group.
Acknowledgements:
The instructor gladly acknowledges other professors for making their course materials available.
