Computer Science 674 - Intelligent Visual Computing

The course will teach students algorithms that intelligently process, analyze and generate visual data. The course will start by covering 2D image and 3D shape representations, classification and regression techniques, and the fundamentals of deep learning. The course will then provide an in-depth background on analysis and synthesis of images and shapes with deep learning, in particular convolutional neural networks, recurrent neural networks, memory networks, auto-encoders, adversarial networks, reinforcement learning methods, and probabilistic graphical models. Particular emphasis will be given on 3D deep learning, an emerging topic in 3D computer vision, graphics and augmented reality.

Open to Computer Science graduate students only. PREVIOUSLY COMPSCI 690IV. MEETS WITH COMPSCI 574. CMPSCI GRADUATES AND OTHERS ARE ASSUMED TO HAVE SUCCESSFULLY COMPLETED PREREQS OR EQUIV: COMPSCI 311, 383, AND 373 (OR 473), ALL WITH A GRADE OF A- OR BETTER. SEATS HELD FOR INCOMING GRAD STUDENT REGISTRATION. STUDENTS NEEDING SPECIAL PERMISSION MUST REQUEST OVERRIDES VIA ON-LINE FORM: https://www.cics.umass.edu/overrides.