Motion Analysis for Human Interaction Detection Using Optical Flow on Lattice Superpixels

作     者：ZHENG Peng CAO Yu WANG Song 

作者机构：School of Computer Wuhan University The Key Laboratory of Aerospace Information Security and Trust Computing Ministry of Education Department of Computer Science and Engineering University of South Carolina 

出 版 物：《Wuhan University Journal of Natural Sciences》 (武汉大学学报（自然科学英文版）)

年 卷 期：2013年第18卷第2期

页      面：109-116页

学科分类：08[工学] 081201[工学-计算机系统结构] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Supported by the National Natural Science Foundation of China(61272453) 

主　　题：superpixel optical flow interaction video 

摘      要：We develop a new video-based motion analysis algorithn to determine whether two persons have any interaction in their meet- ing. The interaction between two persons can be very general, such as shaking hands, exchanging objects, and so on. To make the motio~ analysis robust to image noise, we segment each video flame into a set of superpixels and then derive a motion feature and a motion pattern for each superpixel by averaging the optical flow within the superpixe Specifically, we use the lattice cut to construct the superpixels, which are spatially and temporally consistent across frames. Based on the motion feature and the motion pattern of the superpixels, we develop an algorithm to divide an input video sequence into three consecutive periods： 1） two persons walking toward each other, 2） two persons meeting each other, and 3） two persons walking away fi＇om each other. The experiment show that the proposed algorithm can accurately dis- tinguish the videos with and without human interactions.