Crowd motion segmentation and behavior recognition fusing streak flow and collectiveness

Mingliang Gao; Jun Jiang; Jin Shen; Guofeng Zou; Guixia Fu

doi:10.1117/1.OE.57.4.043109

24 April 2018 Crowd motion segmentation and behavior recognition fusing streak flow and collectiveness

Mingliang Gao, Jun Jiang, Jin Shen, Guofeng Zou, Guixia Fu

Author Affiliations +

Optical Engineering, Vol. 57, Issue 4, 043109 (April 2018). https://doi.org/10.1117/1.OE.57.4.043109

Abstract

Crowd motion segmentation and crowd behavior recognition are two hot issues in computer vision. A number of methods have been proposed to tackle these two problems. Among the methods, flow dynamics is utilized to model the crowd motion, with little consideration of collective property. Moreover, the traditional crowd behavior recognition methods treat the local feature and dynamic feature separately and overlook the interconnection of topological and dynamical heterogeneity in complex crowd processes. A crowd motion segmentation method and a crowd behavior recognition method are proposed based on streak flow and crowd collectiveness. The streak flow is adopted to reveal the dynamical property of crowd motion, and the collectiveness is incorporated to reveal the structure property. Experimental results show that the proposed methods improve the crowd motion segmentation accuracy and the crowd recognition rates compared with the state-of-the-art methods.

Citation Download Citation

Mingliang Gao, Jun Jiang, Jin Shen, Guofeng Zou, and Guixia Fu "Crowd motion segmentation and behavior recognition fusing streak flow and collectiveness," Optical Engineering 57(4), 043109 (24 April 2018). https://doi.org/10.1117/1.OE.57.4.043109

Received: 8 November 2017; Accepted: 3 April 2018; Published: 24 April 2018

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available