Instructor:Deva Ramanan (dramanan@ics.uci.edu)
Office: DBH 4072, 842-4893
Office hours: Wednesdays 10:30-12pm or by appointment
Lectures: T,R 12:30-1:50pm PSCB (Phy Sci Class Blg) 140
Final: Friday Dec 12, 10:30-12:30 pm
News:
Project 5 out; note the 1-week due date, and the new grading policy!
Class website in construction
Course description
This class introduces the computational and mathematical problems of computer vision through the umbrella application of computational photography. Computational photography is an emerging field at the intersection of vision, graphics, and increasingly, machine learning. It deals with computational mechanisms for creating, enhancing, and interpreting digital images. The course will describe algorithmic techniques for low-level image understanding, including applications such as geometric reconstruction, warping & registration, denoising & deblurring, and hole-filling & blending. The course will also introduce high-level algorithms for image understanding including object recognition and detection.
Students will acquire knowledge of image formation, including camera optics, imaging transformations, and basic image processing. The course assumes a background knowledge of linear algebra and probability, but will introduce linear least squares, classification, and dynamic programming.
The class will emphasize hands-on implementation of the presented algorithms through numerous (5) project assignments. Students will be encouraged to acquire their own images of indoor and outdoor scenes for the project assignments. To validate the presented material, there will also be a final exam.
Prerequisite:
CS 6D/Mathematics 6D, Mathematics 6G or 3A, Mathematics 2A-B, CS 23. Please see me if you haven't taken these classes.
Textbook:
There is no official textbook. Readings will be taken from various books.
An optional reference is "Computer Vision: A Modern Approach" by Forsyth and Ponce
Grading:
The course will include 5 MATLAB programming assignments worth 15% each and a final exam worth 25%. (New) I will drop the lowest score from your programming assignments. Hence, only 4 out of the 5 scores will count toward your 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. In particular, much of this course is based on a wonderful set of slides from Fredo Durand, Alyosha Efros, Rob Fergus, Bill Freeman, Steve Seitz, David Martin, and Rick Szeliski.
