The PASCAL Visual Object Classes Challenge Workshop 2008

17th October 2008, ECCV 2008, Marseille, France


This workshop presented and discussed the results of the PASCAL Visual Object Classes Challenge (VOC2008). As in previous years' challenges there are two main competitions, one testing image classification ("does the image contain an instance of this class?"), and one testing object detection ("provide a bounding box for each instance of the class, if any"). In addition there are two 'taster' competitions: the first evaluates the object layout in more detail ("detect the hands, feet etc for a person"), the second evaluates object segmentation at the pixel level.

A new database has been prepared consisting of 20 classes with around 25000 annotated instances in total. The images are obtained from flickr. The classes include people, cats, dogs, cars, motorbikes, bottles and sofas. The annotation includes a rectangular bounding box and flags to indicate pose and level of difficulty.

Full details of the challenge.


The workshop was held on the morning of October 17th 2008, in association with ECCV 2008.


09:00 - 09:25 Overview and results of the classification challenge
Mark Everingham
University of Leeds
09:25 - 10:00 Classification methods

SurreyUVA_SRKDA method
Muhammad Atif Tahir1, Koen van de Sande2, Jasper Uijlings2, Fei Yan1, Xirong Li2, Krystian Mikolajczyk1, Josef Kittler1, Theo Gevers2, Arnold Smeulders2
1University of Surrey; 2University of Amsterdam
[PDF] [Code]

UIUC_CMU method
Derek Hoiem1, Santosh Divvala2, James H. Hays2
1University of Illionois Urbana-Champaign; 2Carnegie Mellon University
10:00 - 10:20 Overview and results of the detection challenge
Mark Everingham
University of Leeds
10:20 - 10:55 Detection methods

LEAR_PlusClass method
Hedi Harzallah, Cordelia Schmid, Frederic Jurie, Adrien Gaidon
INRIA Rhone-Alpes

UoCTTIUCI method
Pedro Felzenszwalb1, Ross Girshick1, David McAllester2, Deva Ramanan3
1University of Chicago; 2TTI Chicago; 3University of California, Irvine
[PDF] [Code]
10:55 - 11:25 Break
11:25 - 11:40 Overview and results of the segmentation taster challenge
John Winn
Microsoft Research Cambridge
11:40 - 12:15 Segmentation methods

XRCE_Seg method
Gabriela Csurka, Florent Perronnin, Yan Liu
Xerox Research Centre Europe (XRCE), Textual and Visual Pattern Analysis Group

BrookesMSRC method
Lubor Ladicky1, Phil Torr1, Pushmeet Kohli2
1Oxford Brookes University; 2Microsoft Research Cambridge
12.15 - 12.30 Conclusions and discussion



The preparation and running of this challenge is supported by the EU-funded PASCAL Network of Excellence on Pattern Analysis, Statistical Modelling and Computational Learning.