A study of human activity sequence segmentation based on gaussian model

Abstract

Human activity analysis based on video is a hot research topic in the field of computer vision, in which segmentation of human activity sequence is a fundamental question. In this paper, we present a new unsupervised segmentation method for human activity sequence. The main idea of this method is that the most dramatic point, which is regarded as the split point, is detected by gaussian model according to human action's mutation in the video. Experimental results show the feasibility and effectiveness of the segmentation algorithm in this paper.