Previous offerings: [Fall04] [Fall05] [Fall06] [Fall07]
Human beings are adept at a variety of visual "recognition" problems, such as recognizing everyday objects or the actions or identities of people. In this course we will explore computer vision techniques that have the grand goal of giving computers these same abilities. Classical techniques and recent original research addressing these problems will be presented.
This course is an 800-level seminar-style course. The goal is to introduce students to important problems and approaches in recognition, especially object recognition and looking at people challenges. This will be accomplished by reading research papers on these topics. For each topic, I will present important papers describing methods for attacking the problems. Students will then present recent research papers building on these methods. The semester's work will culminate in a course project in which students will gain in-depth knowledge about solving a recognition problem.