HEAT Image Retrieval System
HEAT is an image retrieval web-application that is intended for large unstructured collections of images without semantic annotations. The system implements a novel searching paradigm that does not require any explicit query. At each iteration, the system displays a small set of images and the user chooses the image that best matches what she is looking for. After a few iterations, the sets of displayed images are gradually concentrated on images that satisfy the user.
This page contains the source code of the HEAT image retrieval web-application. The code can be used to reproduce the experimental results presented in the papers mentioned below. If you use this source, please cite one of those papers.
1. Iterative Relevance Feedback with Adaptive Exploration/Exploitation Trade-off , Nicolae Suditu and François Fleuret (Bibtex).
2. HEAT: Iterative Relevance Feedback with One Million Images, Nicolae Suditu and François Fleuret (BibTeX).
A demo web-server is available on-line at http://imr.idiap.ch.
You can download the archive here: heat-2.0.tar.gz (6.68MB).
To run the application: extract the archive and follow the instructions in the README file.
The source code is copyright Idiap Research Institute and provided under the GNU Affero General Public License version 3 (AGPLv3).
Please contact Nicolae Suditu for comments and bug reports.