Boosted learning in dynamic Bayesian networks for multimodal detection

T. Chaodhury1, J.M. Rehg2, V. Pavlovic3, A. Pentland1
1Department of Media Arts and Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA
2College of Computing, Georgia Institute of Technology, Atlanta, GA, USA
3Bioinformatics Program, Boston University, Boston, MA, USA

Tóm tắt

Bayesian networks are an attractive modeling tool for human sensing, as they combine an intuitive graphical representation with efficient algorithms for inference and learning. Temporal fusion of multiple sensors can be efficiently formulated using dynamic Bayesian networks (DBNs) which allow the power of statistical inference and learning to be combined with contextual knowledge of the problem. Unfortunately, simple learning methods can cause such appealing models to fail when the data exhibits complex behavior We first demonstrate how boosted parameter learning could be used to improve the performance of Bayesian network classifiers for complex multimodal inference problems. As an example we apply the framework to the problem of audiovisual speaker detection in an interactive environment using "off-the-shelf" visual and audio sensors (face, skin, texture, mouth motion, and silence detectors). We then introduce a boosted structure learning algorithm. Given labeled data, our algorithm modifies both the network structure and parameters so as to improve classification accuracy. We compare its performance to both standard structure learning and boosted parameter learning. We present results for speaker detection and for datasets from the UCI repository.

Từ khóa

#Bayesian methods #Inference algorithms #Face detection #Motion detection #Humans #Sensor fusion #Learning systems #Skin #Mouth #Detectors

Tài liệu tham khảo

10.1109/CVPR.1999.784617 10.1109/ACV.1996.572043 pavlovie?, 2002, Boosting Distributions Ht Snowbird Learning Workshop 10.1109/CVPR.1997.609401 koller, 1997, Object-oriented bayesian networks, Proc of the 13th Conf on Uncertainty in Al, 302 10.1109/CVPR.2000.854730 intille, 1998, Representation and visual recognition of complex, multi-agent actions using belief networks, CVPR Workshop Interpretation of Visual Motion jensen, 1996, An Introduction to Bayesian Networks 10.1109/AFGR.2000.840663 heckerman, 1995, A tutorial on learning with Bayesian networks 10.1023/A:1007614523901 10.1109/ICPR.2002.1048137 10.1109/CVPR.1996.517075 10.1109/CVPR.1997.609450 friedman, 1998, Learning the structure of dynamic probabilistic networks, Proc Uncertainty in AI Conf (UAI 00) blake, 1998, UCI repository of machine learning databases fisher iii, 2000, Learning joint statistical models for audio-visual fusion and segregation, Proc Advances in Neural Information Processing Systems cutler, 2000, Look who's talking: Speaker detection using video and audio correlation, Proc IEEE Int Conf Multimedia and Expo (ICME), 10.1109/ICME.2000.871073 10.1007/BF00994110 10.1145/274644.274668 friedman, 2000, Being bayesian about network structure, Proc 15th Conf Uncertainty in AI 10.1023/A:1007465528199