n-Grams of Action Primitives for Recognizing Human Behavior

摘要

This paper presents a novel approach for behavior recognition from video data. A biologically inspired action representation is derived by applying a clustering algorithm to sequences of motion images. To obey the temporal context, we express behaviors as sequences of n-grams of basic actions. Novel video sequences are classified by comparing histograms of action n-grams to stored histograms of known behaviors. Experimental validation shows a high accuracy in behavior recognition.