TR-IIS-07-008    Fulltext

Visual analysis of human behavior has generated considerable interest in the field of computer vision because it has a wide spectrum of potential applications. A human behavior analysis system must address three main tasks: object detection, human tracking, and understanding behavior. In this paper, we propose a language modeling framework for handling the third task. The framework is comprised of two modules: a posture labeling module, and an atomic action learning and recognition module. A posture template selection algorithm is developed based on a modified shape context matching technique. The posture templates form a codebook that is used to convert input posture sequences into training symbol sequences or recognition symbol sequences. Finally, a variable-length Markov model technique is applied to learn and recognize the input symbol sequences of atomic actions. Experiments on real data demonstrate the efficacy of the proposed system.