Pokemon Go puts Pikachu on the ground, but why does it look 'floaty'?
Snapchat face swap creates funny portraits, but can we make it look real?
HoloLens lets us integrate virtual objects with the world, but can they be indistinguishable from reality?
Visual computing often relies on computer vision to analyse real-world captured imagery and create new natural interactions with computers. In this seminar course, we will discover the state of the art algorithmic contributions in computer vision which make the interactive merging of real-world imagery and graphics possible. Each week, we will read pairs of complementary papers, present them, and discuss their contributions, impact, and limitations. Then, we will build upon our analyses and discussions to create group projects which try to further the state of the art. Beyond computer vision, this course will help you learn how to quickly interpret and assess academic papers, how to help generate research ideas, and how to give effective and engaging presentations.
- We will read two papers per week, and everyone will submit two+ questions for discussion.
(3-6 hours per week)
- In class, we will present the papers. Each student will present at least once, and depending on enrollment, this could be individually or in groups.
(8-10 hours prep; 30-45 minutes presentation)
- After the presentation, together we will discuss the papers in detail to understand them and to generate new ideas. Each student will act as the discussion moderator at least once.
- In groups, we will develop these ideas into a project to try to consider and extend the state of the art.
(4 weeks; 10-14 hours per week; self-determined groups and projects). This project will have a written proposal (5 hours), a written progress update (2 hours), a written report (10 hours), and be shown and discussed in a demo session (3 hours).
- Grade: 30% presentation, 30% discussion, 40% project.
- Instructor help is available at all junctions!
- Time commitment: Over 13 weeks, students will spend 2+ hours per week in class (26 hours). Required reading is expected to take 6 hours per topic (60 hours). Each student presentation is expected to take 10 hours to prepare. Project work is expected to take 56 hours, plus 20 hours for proposals, updates, reports, and demo preparation (76 hours). 8 hours are allotted for outside-class discussion, office hour attendance, and discussion question preparation. 180 hours total.
Please join us!