Location: Computer Sciences 1221
Time: Tues, Thurs 2:30-3:45pm
Credits: 3
Instructor: Yong Jae Lee
Email: yongjaelee@cs.wisc.edu (email subject should begin with "[CS 839]")
Office hours: By appointment
TA: Anirudh Sundara Rajan
Email: asundararaj2@wisc.edu (email subject should begin with "[CS 839]")
Office hours: By appointment
Announcements
(9/7) Please read this website and the detailed course requirements and grading criteria very carefully.
Course Overview
This graduate course introduces students to machine learning based synthesis and manipulation of visual data (images and videos). Both classical (e.g., nearest neighbor, filtering) and modern deep learning based (e.g., ConvNets, GANs, Diffusion Models) algorithms will be presented for image representation, synthesis, and manipulation. It will be a mix of instructor led lectures and student led presentations. Students will learn to implement their own algorithms for image synthesis and manipulation, to understand and analyze state-of-the-art techniques, and to identify interesting open questions and future directions. It should be of relevance to students interested in computer vision, machine learning, and graphics.
Prerequisites
It is recommended that students have taken either an introductory computer vision or machine learning course. Students are also strongly recommended to have basic knowledge of probability and linear algebra, and Python programming experience. Programming will be required for the final project. Please talk to me if you are unsure if the course is a good match for your background.
Requirements
Students will be responsible for completing two problem sets during the first half of the course, and for the second half of the course, writing paper reviews each week, participating in discussions, presenting once/twice in class (depending on enrollment), and completing a final project.
Canvas
We will use Canvas for assignment, paper review, and project proposal/report submissions and grading. Our class page: https://canvas.wisc.edu/courses/375820
Grading
The final grade will be determined by:
Problem sets (20%)
Paper reviews (20%)
Class participation (10%) (Only for paper presentation days)
Paper presentation (20%)
Final project (30%)
Important Dates (tentative)
10/2: Problem Set 1 (PS1) due
10/20: Problem Set 2 (PS2) due
10/13: Final project proposal due
12/5-12/12: Final project presentations
12/13: Final project report due
Detailed course requirements and grading are here.