Best Papers
| CSC 295 |
Computer Vision |
Spring 2010 |
Introduction
We will read and discuss in class one or two of the best papers from
the most recent top vision conferences. In this way, we'll be learning
together:
- How to read research papers
- About the latest in vision research
- What the community thinks is currently important
Candidates
Our candidates (listed in no particular order) are drawn from
CVPR 2009, ECCV 2008, and ICCV 2007. See
the list of papers below and read their abstracts.
-
Single
Image Haze Removal Using Dark Channel Prior, by Kaiming He,
Jian Sun, and Xiaoou Tang (CVPR '09 Best paper).
-
Nonparametric
Scene Parsing: Label Transfer via Dense Scene Alignment, by Ce
Liu, Jenny Yuen, Antonio Torralba (CVPR '09 Best Student Paper)
-
Learning Spatial Context: Using Stuff to Find Things, by Geremy Heitz and Daphne Koller. (ECCV '8 Best Paper)
-
Learning to Localize Objects with Structured Output
Regression, by Matthew B. Blaschko and Christoph
H. Lampert. (ECCV '08 Best Student Paper)
-
Discriminative
Models for Multi-class Object Layout by Chaitanya Desai, Deva
Ramanan and Charless Fowlkes: (ICCV '09 Best Paper)
Voting
Please vote by emailing your RANKED TOP THREE choices (by paper number above) a to the instructor by May 5.
Responses
You will be required to submit an approximately one-page response to
the paper before class to help prepare you for the discussion. In
particular, you should note:
- What problem are they trying to solve?
- Why is the problem important?
- How does it currently get done and what are the limitations?
- What are the authors' goals?
- Does the paper have a scientific thesis? Is it falsifiable?
- What are the paper's claims?
- Are the claims substantiated (by theory or experiment)?
- What are the limitations of the proposed approach?
- Are there ways to extend the method?
You should include at least two primary points that critique,
dispute, extend, or reinforce the paper.
Submit your responses (in plaintext or PDF only) via
P-Web; they are due at the beginnning of class.