Learning to See

(a reading group on machine learning and computer vision)

We meet once in two weeks and present/discuss papers or topics 
related to machine learning and pattern recognition with an 
emphasis on computer vision applications. We focus on papers 
from conferences like NIPS, CVPR, ICML, ICCV, ECCV and journals 
like JMLR, PAMI (just to name a few). The meetings are informal!

Meetings
Day: Monday
Time: 1700 hrs. (lasting around 90 mins.)
Place: 57-568

If you would like to propose a paper, send a mail to Adrian Ulges.

2006 schedule

Date	Paper/Topic	Conference/Journal	Presenter
25.01	Matching Words and Pictures	JMLR 2003	Shankar Vembu
01.02	Discovering object categories in image collections	MIT-CSAIL-TR and ICCV 2005	Adrian Ulges
14.02	Training invariant support vector machines	ML 2002	Daniel Keysers
28.02	Probabilistic graphical models	-	Shankar Vembu
13.03	Monte Carlo Sampling and Applications in Bayesian Networks	-	Adrian Ulges
11.04	Clustering methods	-	Wei Lu
08.05	Musical Alignment	ICASSP 2006	Hagen Kaprykowsky
29.05	Distance Measures for Image Segmentation Evaluation	EURASIP JASP 2006	Faisal Shafait
19.06	Topic models for word completion	ICPR 2006, FinTAL 2006	Elisabeth Wolf
...	...	...	...
05.10.	Introduction to HMMs	---	Oliver Wirjadi, Adrian Ulges

List of papers
The following is a sample list of papers that you might consider presenting.
Video epitomes, CVPR 2005.
Robust Real-time Object Detection, IJCV 2002.
Object Class Recognition Using Discriminative Local Features, PAMI 2004.
Statistics of natural image categories, Network: Comput. Neural Syst. 2003.
Normalized cuts and image segmentation, PAMI 2000.
Conditional Random Fields for Object Recognition, NIPS 2004.
Laplacian Eigenmaps for dimensionality reduction and data representation, Neural Computation, 2003.
Sharing visual features for multiclass and multiview object detection, PAMI 2005.
Using the Forest to See the Trees:A Graphical Model Relating Features, Objects and Scenes, NIPS 2003.
Video Google: A Text Retrieval Approach to Object Matching in Videos, ICCV 2003.
Object Class Recognition by Unsupervised Scale-Invariant Learning, CVPR 2003.
Image Parsing: Unifying Segmentation, Detection, and Recognition, ICCV 2003.
A Probabilistic Approach to Object Recognition Using Local Photometry and Global Geometry, ECCV 1998.

Useful links
Variational Bayes
Recognizing and Learning Object Categories
Object detection bibliography
Computer vision homepage
Machine learning reading groups
Reading list on Bayesian statistics
ML classnotes
Graphical models (with a list of papers at the end)
Pascal challenge
TRECVID Video Retrieval Evaluation
more papers