Abstract
In this paper, we propose the use of audio visual features for person identity verification instead of traditional passwords and pins for secure access. Experiments performed on different gender specific subsets of data from VidTIMIT and UCB N multimodal databases under clean and noisy conditions show that with audio-visual fusion, about 25-40% improvement in performance can be achieved with multimodal fusion as compared to single mode-audio only or visual only mode even under noisy acoustic conditions.
Original language | English |
---|---|
Title of host publication | Proceedings of the 3rd Indian International Conference on Artificial Intelligence, Pune, India, December 17-19, 2007 |
Subtitle of host publication | IICAI 2007 |
Pages | 1-10 |
Number of pages | 10 |
Publication status | Published - 2007 |
Event | 3rd Indian International Conference on Artificial Intelligence, IICAI 2007 - Pune, India Duration: 17 Dec 2007 → 19 Dec 2007 |
Conference
Conference | 3rd Indian International Conference on Artificial Intelligence, IICAI 2007 |
---|---|
Country/Territory | India |
City | Pune |
Period | 17/12/07 → 19/12/07 |