A Bayesian Computer Vision System for Modeling Human Interaction

We describe a real-time computer vision and machine learning system for modeling and recognizing human behaviors in a visual surveillance task. The system is particularly concerned with detecting when interactions between people occur, and classifying the type of interaction. Examples of interesting interaction behaviors include following another person, altering one's path to meet another, and so forth. Our system combines top-down with bottom-up information in a closed feedback loop, with both components employing a statistical Bayesian approach. We propose and compare two different state-based learning architectures, namely HMMs and CHMMs, for modeling behaviors and interactions. The CHMM model is shown to work much more efficiently and accurately. Finally, to deal with the problem of limited training data, a synthetic 'Alife-style' training system is used to develop flexible prior models for recognizing human interactions. We demonstrate the ability to use these a priori models to accurately classify real human behaviors and interactions with no additional tuning or training.

[1]  Geoffrey E. Hinton,et al.  Mean field networks that learn to discriminate temporally distorted strings , 1991 .

[2]  Hans-Hellmut Nagel,et al.  From image sequences towards conceptual descriptions , 1988, Image Vis. Comput..

[3]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[4]  Aaron F. Bobick,et al.  Computers Seeing Action , 1996, BMVC.

[5]  Alex Pentland,et al.  Probabilistic visual learning for object detection , 1995, Proceedings of IEEE International Conference on Computer Vision.

[6]  Wray L. Buntine.  Operations for Learning with Graphical Models , 1994, J. Artif. Intell. Res..

[7]  Jitendra Malik,et al.  Automatic Symbolic Traffic Scene Analysis Using Belief Networks , 1994, AAAI.

[8]  Michael I. Jordan,et al.  Boltzmann Chains and Hidden Markov Models , 1994, NIPS.

[9]  Alex Pentland.  Classification by Clustering , 1976 .

[10]  Alex Pentland,et al.  Coupled hidden Markov models for complex action recognition , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[11]  Alex Pentland,et al.  LAFTER: lips and face real time tracker , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  H. Buxton,et al.  Advanced visual surveillance using Bayesian networks , 1997 .

[13]  Alex Pentland,et al.  Graphical Models for Recognizing Human Interactions , 1998, NIPS.

[14]  A. Pentland,et al.  Blob - An unsupervised clustering approach to spatial preprocessing of MSS imagery , 1977 .

[15]  Anthony G. Cohn,et al.  Building qualitative event models automatically from visual input , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[16]  Alex Pentland,et al.  Pfinder: Real-Time Tracking of the Human Body , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Alex Pentland,et al.  Modeling and Prediction of Human Behavior , 1999, Neural Computation.

[18]  Wray L. Buntine.  A Guide to the Literature on Learning Probabilistic Networks from Data , 1996, IEEE Trans. Knowl. Data Eng..

[19]  Michael I. Jordan,et al.  Probabilistic Independence Networks for Hidden Markov Probability Models , 1997, Neural Computation.

[20]  Robert C. Bolles,et al.  The Representation Space Paradigm of Concurrent Evolving Object Descriptions , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Alex Pentland,et al.  Active gesture recognition using partially observable Markov decision processes , 1996, Proceedings of 13th International Conference on Pattern Recognition.