A Generative Approach to Audio-Visual Person Tracking
This paper focuses on the integration of acoustic and visual information for people tracking. The system presented relies on a probabilistic framework within which information from multiple sources is integrated at an intermediate stage. An advantage of the method proposed is that of using a generative approach which supports easy and robust integration of multi source information by means of sampled projection instead of triangulation. The system described has been developed in the EU funded CHIL Project research activities. Experimental results from the CLEAR evaluation workshop are reported.
KeywordsSound Source Coherence Measure Microphone Array Target Height Coherent Noise
Unable to display preview. Download preview PDF.
- 1.CLEAR 2006 evaluation campaign. http://www.clear-evaluation.org/
- 2.Rich transcription 2005 spring meeting recognition evaluation. http://www.nist.gov/speech/tests/rt/rt2005/spring/
- 3.Brandstein, M., Ward, D.: Microphone Arrays. Springer, Heidelberg (2001)Google Scholar
- 4.Comaniciu, D., Ramesh, V., Meer, P.: Real-time tracking of non-rigid objects using mean-shift. In: Int. Conf. on Computer Vision and Pattern Recognition, vol. 2, pp. 142–149 (2000)Google Scholar
- 5.De Mori, R.: Spoken Dialogue with Computers. Academic Press, London (1998)Google Scholar
- 8.Isard, M., MacCormick, J.: BraMBLe: A Bayesian multiple-blob tracker. In: Int. Conf. of Computer Vision, vol. 2, pp. 34–41 (2003)Google Scholar
- 9.Lanz, O.: Approximate bayesian multibody tracking. IEEE Trans. Pattern Analysis and Machine Intelligence, to appear (2006)Google Scholar
- 10.Omologo, M., Svaizer, P.: Acoustic event localization using a crosspower-spectrum phase based technique. In: Int. Conf. on Acoustics, Speech, and Signal Processing, vol. 2, pp. 273–276 (1994)Google Scholar
- 11.Sullivan, J., Rittscher, J.: Guiding random particles by deterministic search. In: Int. Conf. of Computer Vision, vol. 1, pp. 323–330 (2001)Google Scholar