The goal of the student-led lectures is for you to develop your skills in analyze and present cutting-edge research in computer vision. Each student/team will lead one week's lectures on one computer vision topic.
This effort will constitute 30% of your total class grade. Your grade for this effort will be calculated as follows:
For the paper proposal, you and your teammate you should:
Identify 4-6 candidate papers that were recently published at a premiere computer vision conference (e.g., CVPR, ICCV, ECCV) on your topic. Two of these papers will be assigned as required readings, with one needing to be about a specific dataset challenge and the other about a computer vision model.
Select a 30 minute time slot that works for all group members to meet with the instructor to discuss which papers to cover during the presentation. Available time slots will be posted via Canvas. This meeting must be at least two weeks prior to your first lecture.
Email the candidate papers to the instructor at least 72 hours in advance of the meeting.
Presentation Review
The presentation review is a chance to review the slide decks for both lectures and resolve any open questions. Expected content for each lecture is outlined in the next two sections. For this presentation review, you will be expected to:
Select a 30 minute time slot that works for all group members to meet with the instructor to review the lecture slides for both lectures. Available time slots will be posted via Canvas. This meeting must be at least one week prior to your first lecture.
Email the lecture slides to the instructor at least 72 hours in advance of the meeting.
Lecture 1
The first lecture should include a 45-minute presentation that: (i) defines the problem, (ii) motivates the practical importance of solving this problem with a computer vision solution (i.e., applications that can/do benefit society), (iii) describes 1-2 existing datasets used to track progress on this problem, and (iv) describes metric(s) used to evaluate the performance of computer vision models. Then, the lecture will conclude with a facilitated class discussion about the merits and limitations of existing community-shared datasets and evaluation metrics, organized by the instructor around the questions and discussion points submitted by all students. If materials from outside sources are included in the presentations (for example, specific figures or slides), proper credit MUST be given.
Lecture 2
The second lecture should include a 45-minute presentation that describes at least two papers that each introduce a computer vision model. The presentation should cover: (i) the novelty claims of each paper, (ii) mechanisms used to validate the claims, and (iii) open technical questions/problems. Optionally, this lecture can also include a programming tutorial and/or demo. Then, the lecture will conclude with a facilitated class discussion about the merits and limitations of existing computer vision models, organized by the instructor around the questions and discussion points submitted by all students. If materials from outside sources are included in the presentations (for example, specific figures or slides), proper credit MUST be given.