Behavior Recognition in Human Object Interactions with a Task Model

Behavior recognition can be greatly levitated by a model of the task with which the behavior is associated. We present a Markovian task model that captures the temporal relations between subtasks, and provides prior information that helps behavior recognition from simple and noisy sensory data. A Dynamic Bayesian Network is used to integrate multi-model evidence and infer underlying task states. Experiments demonstrate that this system can recognize human behavior in everyday tasks such as making sandwiches.

[1]  A. L. Yarbus,et al.  Eye Movements and Vision , 1967, Springer US.

[2]  Alex Pentland,et al.  Towards perceptual intelligence: statistical modeling of human individual and interactive behaviors , 2000 .

[3]  Context-Aware Computing,et al.  Inferring Activities from Interactions with Objects , 2004 .

[4]  Michael J. Swain,et al.  Color indexing , 1991, International Journal of Computer Vision.

[5]  A. L. I︠A︡rbus Eye Movements and Vision , 1967 .

[6]  Dana H. Ballard,et al.  Routine based models of anticipation in natural behaviors , 2005, AAAI 2005.

[7]  Andrew Liu,et al.  MODELING AND PREDICTION OF HUMAN DRIVER BEHAVIOR , 2001 .

[8]  Chen Yu,et al.  Learning to recognize human action sequences , 2002, Proceedings 2nd International Conference on Development and Learning. ICDL 2002.