Robust Eye Detection under Active Infrared Illumination
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 04
The BANCA database and evaluation protocol
AVBPA'03 Proceedings of the 4th international conference on Audio- and video-based biometric person authentication
Hi-index | 0.00 |
Audio- and video-based person authentication has gained significant attention. Common databases are crucial for comparative evaluation of algorithms. Almost all of the publicly available databases contain images/videos captured in visible illumination. In this paper we present an acquisition system and procedure for collection of a multi-modal database under near infrared (NIR) illumination. Audio and video data of 74 subjects have been recorded under simulated realistic conditions. Multi-view images with different facial expressions have been captured under strictly controlled conditions. Ground truths, i. e. the manually marked eye positions and lip corners, are also provided. The database is publicly available for research purposes on request from Hamburg University of Technology (http://www.ti1.tu-harburg.de/TUNIR).