Object Recognition

Recognition of objects in cluttered scenes is an interesting and important open problem in computer vision and pattern recognition. In close connection to the development of our content-based image retrieval system FIRE we are focussing on this topic. Some efforts of out work in this area are documented by our CVPR 05 publication and by our results in the PASCAL Visual Object Classes Challenge 2005.

Most of our software for this work is part of the FIRE framework.

Content-based Image Retrieval

FIRE (Flexible Image Retrieval Engine) is our content-based image retrieval system. We have large efforts on examining features, distance functions, ways to measure the retrieval performance of CBIR systems.

FIRE is available under the terms of the GNU General Public license. Further information on FIRE, the source code, and relevant publication can be found here.

An online demonstration of FIRE is available here. The server usually is running. If it is not up, don't hesitate to complain about it. I am happy to make it run again.

Evaluation of Image Retrieval

I am a member of the steering committee of the Cross Language Evaluation Forum (CLEF) where I am mainly involved in the ImageCLEF campaign for the evaluation of image retrieval systems.

I have organised the following tasks in various ImageCLEF campaigns:

Complete Annotation of the UW database

In the course of measuring performance of CBIR systems, we created the annotation of the not yet annotated images of the UW database. Information on this, the software used for annotation, and the annotation itself is available here.

Clustering of Images

In the course of creating a CBIR system, a program to cluster images based on various features and distance measures was created as well. It is part of the old FIRE system.

Modelling deformation in images

Here you find our W2D image deformation and recognition software

Last modified: Tue Nov 6 11:09:53 CET 2007

Valid HTML 4.01! xemacs logo debian logo