International Workshop on Object Recognition

October 10-12 2004

Objectives

The ability to recognize living creatures and inanimate objects in photographs or video clips is a critical enabling technology for a wide range of applications including defense, health care,  human-computer interaction, image retrieval and data mining, industrial and personal robotics, manufacturing, scientific image analysis, space exploration, surveillance and security, and transportation. In fact, with the ever expanding array of imagery sources, some form of automatic object recognition technology must eventually be an integral part of every information system.  Despite 40 years of research, however, today's recognition systems are still largely unable to handle the extraordinarily wide range of appearances assumed by common objects in typical images.
The tenet of this workshop is that fundamental new advances in automated object recognition can be achieved by integrating the sophisticated geometric and physical image formation models developed in the computer vision community with the effective models of data distribution and classification procedures developed in the statistical learning theory and theoretical computer science communities.
This three-day workshop will bring together prominent computer vision, machine learning, and computational geometry researchers interested in the fundamental and applicative aspects of object recognition, as well as representative of industry and funding agencies. Its goals are (1) to promote the creation of an international object recognition community, with common datasets and evaluation procedures, (2) to map the state of the art and identify the main open problems and opportunities for synergistic research, and (3) to articulate the industrial and societal needs and opportunities for object recognition research worldwide.

Participants

· Yali Amit (U. Chicago)

· Kobus Barnard (University of Arizona)

· Chris Bishop (Microsoft)

· Joachim Buhmann (Bonn)

· Stefan Carlsson (Royal Institute of Technology)

· Tim Cootes (Manchester)

· Chris Dance (Xerox)

· Nando de Freitas (UBC)

· Pinar Duygulu (Ankara)

· Boris Ephstein (Weizmann Inst.)

· Vittorio Ferrari (ETHZ)

· Olivier Faugeras (INRIA)

· Rob Fergus (Oxford)

· Bill Freeman (MIT)

· Christophe Garcia (France Telecom R&D)

· Don Geman (John Hopkins)

· Martial Hebert (Carnegie Mellon University)

· Geoff Hinton (Toronto)

· David Hogg (Leeds)

· Anthony Hoogs (GE)

· Dan Huttenlocher (Cornell)

· Geoff Hinton (Toronto)

· Michael Isard (Microsoft)

· Frederic Jurie (INRIA)

· Ken'ichi Kitahama (Toyota)

· Jan Koenderink (Utrecht)

· Christophe Laurent (France Telecom R&D)

· Svetlana Lazebnik (University of Illinois)

· Yann LeCun (NYU)

· Fei-Fei Li (Caltech)

· David Lowe (UBC)

· Jitendra Malik (Berkeley)

· Kevin Murphy (UBC)

· Margarita Osadchy (Technion)

· Pietro Perona (Caltech)

· Jean Ponce (University of Illinois)

· Jim Rehg (Georgia Tech)

· Henri Sanson (France Telecom)

· Cordelia Schmid (INRIA)

· Bernhard Schoelkopf (MPI for Biological Cybernetics)

· Bernt Schiele (Darmstad)

· Tom Strat (DARPA)

· Rahul Sukthankar (Intel)

· Rick Szeliski (Microsoft)

· Phil Torr (Oxford Brookes University)

· Antonio Torralba (MIT)

· Bill Triggs (CNRS)

· Michalis Titsias (Edinburgh)

· Akihiro Tsukada (Toyota)

· Shimon Ullman  (Weizmann Institute)

· Allan Yuille (UCLA)

· Luc Van Gool (Catholic University of Leuven)

· John Winn (Microsoft)

· Andrew Zisserman (Oxford)

Sponsors

• Institut National de Recherche and Informatique and Automatique
• PASCAL Network
• DARPA
• General Electric
• Lockheed Martin
• Microsoft
• France Telecom
• Toyota
• Xerox
• Intel

Local Arrangements

• The workshop will be held at Hellenia Yachting Hotel, Giardini Naxos (ME) Italia Via Jannuzzo, 41, Phone: ++39-094251737, Fax: ++39-094254310.
• The nearest airport in Catania (better than Palermo, the other option). Attendees from the U.S. should fly on U.S. carriers. There are three possibilities for transportation from the airport : (1) take a bus from the airport to Taormina stazione (4.7 Euros, 1hr 15mn drive) then a cab from there to Hellenia (about 10 Euros, 5mn drive). The bus runs every 1/2 hour in the morning, then every hour (9:45, 10:45, etc.) after that. (2) take a cab from the airport to Hellenia Yachting Hotel. It Should cost about 85 Euros, and take about 45mn to 1hr. (3) rent a car at the airport.

Abstracts

Program

Sunday October 10
9.00-9.15	Introduction Jean Ponce
Recognition I
9.15- 9.45	A Concise Taxonomy of Recognition Pietro Perona
9.45-10.15	Scale-Invariant Object Categorization and Segmentation using a Scale-Adaptive Mean-Shift Search Algorithm Bernt Schiele
10.15-10.45	Layered Pictorial Structures Phil Torr
10.45-11.00	Break
11.00-11.30	Statistical Relational Models for Object Recognition Dan Huttenlocher
11.30-12.00	Evolution of the constellation model Rob Fergus
12.00-12.30	Using Statistical Models of Shape and Appearance for Face Recognition/Verification Tim Cootes
12.30-4.00	Lunch/Break
4.00-4.30	Variations on Image and Shape Warping, Statistics and Segmentation Olivier Faugeras
4.30-5.00	Information Features for Object Detection Alan Yuille
5.00-5.30	How to Invert Computer Graphics Geoff Hinton
5.30-5.45	Break
5.45-6.15	Contextual Models for Multiclass Object Detection Antonio Torralba
6.15-6.45	Sharing Features for Multi-Class Object Detection Bill Freeman
Short Talks
6.45-7.00	Hierarchical Part-Based Model for Visual Object Categorization Bill Triggs
7.00-7.15	Combinatorial Feature Selection Stefan Carlsson
7.15-7.30	Action Classification for Table Top Games David Hogg

Monday October 11
Recognition II
9.00-9.30	Classification and Recognition by a Hierarchy of Extended Fragments Shimon Ullman
9.30- 10.00	Hierarchical Designs for Object Recognition Don Geman
10.00-10.30	Model-Based Object Classification Yali Amit
10.30-10.45	Break
10.45-11.15	Comparison of Affine Covariant Detectors Cordelia Schmid
11.15-11.45	PCA-SIFT: Improving Matching Accuracy for Local Image Descriptors Rahul Sukthankar
11.45-12.15	Integrating Multiple Model Views for Object Recognition Vittorio Ferrari
12.15-12.45	Semi-Local Affine Parts for Object Recognition Svetlana Lazebnik
12.45-4.00	Lunch/Break
4.00-5.30	Panel: Datasets Moderator: Jean Ponce · Andrew Zisserman and Luc VanGool: The Pascal challenge · Chris Bishop: Progress at Microsoft · David Lowe: Testing recognition approaches · Joachim Buhmann: How Far Do Purely Local Features Lead in Object Categorization? · Kobus Barnard: NSF proposal
5.30-5.45	Break
5.45-6.15	Advances in Visual Categorization with Bags of Keypoints Chris Dance
6.15-6.45	Dense Feature Correspondences Make Recognition Easy Jitendra Malik
Short Talks
6.45-7.00	Language Structure and Image Understanding Kobus Barnard
7.00-7.15	Kernel Methods for Implicit Surface Modeling Bernhard Schoelkopf
7.15-7.30	Comparison of Generative and Discriminative Approaches to Object Recognition Chris Bishop

Tuesday October 12
Recognition III
9.00-9.30	Loss Functions for Discriminative Training of Energy-Based Models with Applications to Object Recognition Yann LeCun
9.30- 10.00	Synergistic Face Detection and Pose Estimation with Energy-Based Models Margarita Osadchy
10.00-11.00	Tutorial on Variational Message Passing and Its Applications for Learning Object Category Models Fei-Fei Li and John Winn
11.00-11.15	Break
11.15-12.30	Panel: Industrial Needs, Applications, and Research Futures Moderators: Martial Hebert and Cordelia Schmid · Ken'ichi Kitahama (Toyota) · Tom Strat (DARPA) · Rick Szeliski (Microsoft) · Cris Dance (Xerox) · Anthony Hoogs (GE) · Christophe Garcia (France Telecom)
12.30-4.00	Lunch/Break
4.00-4.30	Recognition as Machine Translation: Labeling Objects and Faces Using Large Image and Video Collections Pinar Duygulu
4.30-5.00	Bayesian Multiple Instance Learning Models for Object Recognition Nando de Freitas
	Pitfalls in Recognizing Panorama Rick Szeliski
5.00-5.15	Break
5.15-5.45	Recognition and Segmentation of Broad Categories of Scene Content Anthony Hoogs
5.45-6.15	3D Object Congruence in Human Pictorial Space (movie version) Jan Koenderink
6.15-8.15	Posters and demonstrations Classification by Feature Hierarchies Boris Ephstein Scale-invariant Shape Features for Recognition of Object Frederic Jurie Natural Image Classification Using Foveal Strings Christophe Laurent Fast Unsupervised Greedy Learning of Multiple Objects and Parts from Video Michalis Titsias Fast and Automatic Induction of Cascade Classifiers Jim Rehg Segmenting, Modeling, and Matching Video Clips Containing Multiple Moving Objects Fred Rothganger, Svetlana Lazebnik, Cordelia Schmid, and Jean Ponce The Challenges and Rewards of Category-Level 3D Object Recognition UIUC Computer Vision and Robotics Group 3D Human Pose from Silhouettes by Relevance Vector Regression Ankur Agarwal and Bill Triggs A Visual Category Filter for Google Images Rob Fergus, Pietro Perona, and Andrew Zisserman
8.30	Banquet