Pedestrian Head Detection and Tracking Using Skeleton Graph for People Counting in Crowded Environments

This paper describes a new head detection method for people counting in crowded environments from a single camera. Our method adopts skeleton graph to distinguish person among people in crowded enviroments. The usage of skeleton graph is the main difference between this method and the traditional ones. Firstly, the skeleton graphs are calculated for each selected blob in the scene after foreground estimation. Then, we explore the structural property of each blob for a head detection and to predict a number of people. Each detected head in a skeleton silhouette is identified as independent state or partially occluded state, and during tracking every state is updated. Finally, the experimental results are shown to demonstrate the robustness of our method.

[1]  Luc Van Gool,et al.  Object Tracking with an Adaptive Color-Based Particle Filter , 2002, DAGM-Symposium.

[2]  Jitendra Malik,et al.  Estimating Human Body Configurations Using Shape Context Matching , 2002, ECCV.

[3]  Hironobu Fujiyoshi,et al.  Real-time human motion analysis by image skeletonization , 1998, Proceedings Fourth IEEE Workshop on Applications of Computer Vision. WACV'98 (Cat. No.98EX201).

[4]  Paul A. Viola,et al.  Detecting Pedestrians Using Patterns of Motion and Appearance , 2005, International Journal of Computer Vision.

[5]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[6]  Gregory D. Hager,et al.  Fast and Globally Convergent Pose Estimation from Video Images , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Robert T. Collins,et al.  Evaluation of sampling-based pedestrian detection for crowd counting , 2009, 2009 Twelfth IEEE International Workshop on Performance Evaluation of Tracking and Surveillance.

[8]  Nuno Vasconcelos,et al.  Privacy preserving crowd monitoring: Counting people without people models or tracking , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Larry S. Davis,et al.  Ghost: a human body part labeling system using silhouettes , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[10]  Nicolas Thome,et al.  Learning articulated appearance models for tracking humans: A spectral graph matching approach , 2008, Signal Process. Image Commun..

[11]  Yongtae Do.  Region Based Detection of Occluded People for the Tracking in Video Image Sequences , 2005, CAIP.

[12]  Yannick Boursier,et al.  Sparsity-driven people localization algorithm: Evaluation in crowded scenes environments , 2009, 2009 Twelfth IEEE International Workshop on Performance Evaluation of Tracking and Surveillance.

[13]  Antonio Albiol,et al.  VIDEO ANALYSIS USING CORNER MOTION STATISTICS , 2009 .

[14]  Jean-Philippe Thiran,et al.  Counting Pedestrians in Video Sequences Using Trajectory Clustering , 2006, IEEE Transactions on Circuits and Systems for Video Technology.

[15]  Mario Vento,et al.  A Method Based on the Indirect Approach for Counting People in Crowded Scenes , 2010, 2010 7th IEEE International Conference on Advanced Video and Signal Based Surveillance.