CSE 252C Fall 2006 (original) (raw)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
UNIVERSITY OF CALIFORNIA, SAN DIEGO

CSE 252C: Selected Topics in Vision & Learning

Students taking the course for four units should follow these project guidelines. Here is the feedback form for presentations. Click here to join the class email-list.

Please read, reflect upon, and follow these presentation guidelines, kindly provided by Prof. Elkan. Immediately after your presentation, please email to sjb+cse252c@cs a copy of your slides. For ease of viewing, please make this copy be two slides per page in Adobe PDF.

The schedule of papers and presentations is below. Participants who have not chosen a paper yet should look at the list of suggested papers and contact the instructor.

If you want to change your presentation date, please arrange a swap with another student and notify the instructor at least two weeks in advance.

SUGGESTED PAPERS

MEETING SCHEDULE

OVERVIEW

CSE 252C is a graduate seminar devoted to recent research on pattern recognition and computer vision.

Students may enroll for one, two, or four units:

• For one unit, a student will present one paper as described below.
• For two units, a student will make two separate presentations. (This option may be eliminated.)
• Four units will require two presentations and a project. The class section id for CSE 252C is #570322.

The course is open to anyone who has already taken at least one graduate course in computer vision, artificial intelligence, or a closely related area. Appropriate courses at UCSD include CSE 250A, CSE 250B, CSE 252AB, CSE 254, CSE 253, CogSci 202, ECE 270A, and CSE 275A.

We will meet on Tuesdays and Thursdays from 2:00pm-3:20pm in U413-1 (University Center). The first meeting will be on Thursday September 21, and the final meeting will be on Thursday November 30, 2006.

Possible topics include:

• grouping & segmentation
• structure from motion
• dimensionality reduction
• graph matching
• spectral graph theory
• support vector machines
• kernel-based methods
• regularization
• expectation-maximization (EM) algorithm
• multidimensional scaling
• particle filtering
• object tracking
• object recognition
• markov chain monte carlo (MCMC) methods
• markov random fields
• boosting

Students are encouraged to investigate both fundamental algorithmic issues as well as application areas such as biometrics, content based image retrieval, texture synthesis, motion capture, and image based rendering.

The instructor is Serge Belongie, Assistant Professor, EBU3b room 4118. Office Hours: Tu 4-5pm, W 4-5pm.

Feel free to send email to sjb+cse252c@cs with any questions.

RELEVANT TEXTS

Pattern Recognition and Machine Learning, Bishop.
Computer Vision: A Modern Approach, Forsyth and Ponce
Introductory Techniques for 3-D Computer Vision Trucco and Verri
An Invitation to 3D Vision: From Images to Geometric Models, Y. Ma, S. Soatto, J. Kosecka, S. Sastry
Multiple View Geometry in Computer Vision by Hartley & Zisserman
The Geometry of Multiple Images by Faugeras, Luong, and Papadopoulo
Vision Science: Photons to Phenomenology by Stephen E. Palmer

SEMINAR ORGANIZATION

Each class meeting of 80 minutes will be divided into two parts. First, a student will give a talk lasting about 60 minutes presenting a recent technical paper in detail. In questions during the talk, and in the final 20 minutes, all seminar participants will discuss the paper and the issues raised by it.

Some papers will be theoretical, and some will be applied. Two related applications papers may be discussed together. Theoretical papers will typically be presented and discussed alone, to ensure that mathematical and algorithmic questions are discussed in sufficient depth.

In the first week, we will make a schedule of papers and presentations for the whole quarter. With 10 participants, each student will make two separate presentations. The procedure for one presentation is as follows:

• Two weeks in advance: Finish a draft of 25 to 30 slides that present clearly the work in the paper. Make an appointment with the instructor to discuss the draft slides. Email the slides to sjb+cse252c@cs.
• Ten days in advance: Meet for 30 to 60 minutes to discuss improving the slides, and how to give a good presentation.
• Day of presentation: Give a good presentation with confidence, enthusiasm, and clarity.
• Immediately afterwards: Make changes to the slides suggested by the class discussion. Email the slides in PDF, two slides per page, to the instructor for publishing.

Presentations will be evaluated, in a friendly way but with high standards. Each presentation should be prepared using LaTeX or Powerpoint. You should copy equations, diagrams, charts, and tables as necessary from the paper for the presentation.

For each presentation, we will have a web-based discussion area. Each seminar participant is expected to contribute at least one message to the discussion, before the presentation. A message may ask an interesting question, point out a strength or weakness of the paper, or answer a question asked by someone else. Messages should be thoughtful!

Each student will also do one term project following specific guidelines. The project should be at the frontier of current research, and preferably closely inspired by one of the papers discussed in the class. Project reports will be evaluated using these grading criteria. There is a schedule for handing in a detailed project proposal, a draft project report, and then the final report.

The seminar will have no final exam. Final grades will be based 50% on presentations and participation in class and in the web-based discussions. The other 50% will be the project report.

Most recently updated on Sept. 6, 2006 by Serge Belongie.