CMU Advanced Perception Seminar, Spring 1998


Table of Contents

Class Format

The Advanced Perception course is a graduate reading seminar, meeting once a week to discuss a set of papers covering a specific topic in computer vision. We will look at historically important papers in field, as well as current papers from recent conferences and journals. By reading a mixture of both types of papers each week, we will be able to trace the development of some the fundamental ideas that make up current-day computer vision research.

Each week, from 5-7 papers on a particular topic will be assigned. You will read all the papers, and choose one or more of them as the subject of a short critique (say 2-3 pages) that briefly summarizes and analyzes the paper(s). The critiques will be handed in for grading. During class, the discussion of each paper will lead off with a short (5-10 minute) presentation by one of the students (one student per paper, assigned the week before) elaborating on their critique of the paper. This presentation will then evolve into a class discussion lasting anywhere from 20 minutes to an hour, depending on the paper. The instructors are responsible for keeping the discussion in a fruitful vein and making sure all students get a chance to participate. The instructors are also responsible for making sure that the important points are touched upon during the discussion, which will sometimes mean asking questions of the class, and for making sure that all the papers are covered (which sometimes means cutting off discussion and moving on). One practice that has worked well in the past is to go around the room a few times during the class and get answers from each person to a specific question, such as `How do you rate this paper on a scale of 1-5?' These questions often spark other discussions and help to involve everyone in the class.

What Should be in a Critique?

Critiques provide a short summary and analysis of the technical content of a paper. Critique writing is an important component of the class, and serves several goals: to give you practice in technical writing, to concretely organize your ideas in preparation for class discussion, and to develop the skills necessary to become a good conference/journal paper referee. Furthermore, getting in the habit of writing critiques of the papers you have read will help you do better research - a good critique provides a concise summary that you can refer to later without having to dig out and read the original work, and can provide a written starting point for the obligatory literature review section of your own papers/thesis. To help provide you with a sense for what goes into a critique, see the handout `The Task of the Referee,' by Alan Jay Smith, particularly the section entitled `Evaluating a Research Paper.'

We have found that it is helpful to us, when grading critiques, to have them all in a consistent format. We ask you to hand in critiques with the following sections:

  1. Citation: the title, author, year, and publication citation of the paper you are reviewing
  2. Reviewer: your name and the date
  3. A one-sentence summary (executive abstract) of the paper
  4. A short overview of the paper including a) key ideas, b) technical approach and c) results.
  5. Strong points of the paper
  6. Weak points of the paper
  7. Analysis/Comparison/Ratings
  8. Questions and Issues

We will grade critiques on a three-level scale: check-minus, check, check-plus. Items 1-6 above are required to attain a check. In items 7 and 8 we are looking for your resourcefulness, initiative, creativity and depth of analysis. Doing well in items 7 and 8 will give you a check-plus. Missing any required sections (1-6) or lack of effort in on of them results in a check-minus.

Pay attention to your spelling and English grammar.

Grading Policy

You will be graded on the following items:
1. Written Critiques(40%)
2. Oral Presentations(20%)
3. Class Participation(20%)
4. Take-Home Final(20%)
5. Extra Credit(10%)
110% total

Written critiques form the highest-weighted category, as they represent the bulk of the work that you will be performing (aside from reading the papers themselves). You will be handing in at least one critique per week, which will be graded based on your demonstration that you know what the paper is about and have carefully considered the technical approach and reported results. If you hand in multiple critiques, the best grade will be recorded and the others will increment your extra credit account. Extra credit will also be awarded for critiques that compare and contrast two or more papers that were read that week.

Oral presentation refers to the short summary of a paper that is given in class to jog people's memory of the paper and lead into the discussion session. It is expected to contain the same information as a critique, but presented orally (using transparencies if you wish). Depending on class size, you will be giving roughly three oral paper presentations during the semester.

Class participation is rather hard to judge objectively (but we are going to try). We highly encourage you to participate in class discussion, and indeed, this type of class will be a complete failure if nobody speaks up with their opinions. On the other hand, we don't wish to penalize folks who aren't naturally talkative. We will try to ensure that even soft-spoken people get a chance to air their opinions, and will attempt to grade based on the insightfulness of your comments, rather than the frequency or volume.

There will be a take-home final exam. It will involve writing!

The extra credit category will reflect both objective evidence and subjective impressions we receive that indicate you are genuinely putting in a lot of effort. Anything you do (of a professional nature, related to this class) that makes us like you better, will increase your extra credit score.

Computer Vision Resources

There are many places to go to look for computer vision papers, ranging from archival journals to on-line web sites. Here is a list of our favorite sources of material:

Archival Journals

Conference Proceedings

WWW Resources

Overview of Topics by Week

Week 1: Introduction and Explanation

Introduction; explanation of class format and logistics; distribution of papers for week 2. Instructors talk about computer vision resources, and why particular papers were selected for this course. Discussion of how to write a good critique.

Week 2. Feature Extraction I: Edge Extraction

Week 3. Feature Extraction II: Region Segmentation

Week 4. Feature Extraction III: Active Contours

Week 5. Object Recognition I: Model-Based Methods

Week 6. Object Recognition II: Appearance-Based Methods

The following two papers were packaged together, for the purposes of critiquing:

Week 7. Projective Geometry I: Object Indexing Using Invariants

The following are background papers on projective geometry. These are not required reading, nor will they be presented or critiqued.

Week 8. Projective Geometry II: Structure From Multiple Views

The following two papers will be treated as one, for the purposes of critiquing:

Here are the remaining papers for this week:

Week 9. Video Analysis I: Stabilization And Mosaicing

Week 10. Video Analysis II: Feature Tracking And Optical Flow

Week 11. Video Analysis III: Recovering Egomotion And Structure

Week 12. New View Synthesis

Week 13. Human-Computer Interaction