Track-based and object-based occlusion for people tracking refinement in indoor surveillance

ABSTRACT

People tracking deals with problems of shape changes, self-occlusions and track occlusions due to other interfering tracks and fixed objects that hide parts of the people shape. These problems are more critical in indoor surveillance and in particular in home automation settings, in which the need to merge information obtained form different cameras distributed around the house calls for the integration of reliable data obtained during time. Therefore, tracking algorithms should be carefully tuned to cope with occlusions and shape changes, working not only at pixel level but also at region level. In this work we provide a novel technique for object tracking, based on probabilistic masks and appearance models. Occlusions due to other tracks or due to background objects and false occlusions are discriminated. The classification of occluded regions of the track is exploited in a selective model update. The tracking system is general enough to be applied with any motion segmentation module, it can track people interacting each other and it maintains the pixel to track assignment even with large occlusions. At the same time, the model update is very reactive, so as to cope with sudden body motion and silhouette's shape changes. Due to its robustness, it has been used in different experiments of people behavior control in indoor situations.

REFERENCES

• 1: D. Beymer, K. Konolige, "Real-time tracking of multiple people using continuous detection", Int. Conf. on Computer Vision, 1999.
• 2: I. Cohen, G. Medioni. "Detecting and Tracking Moving Objects in Video Surveillance" Proc. of the IEEE CVPR 99, Fort Collins, June 1999.
• 3: Rita Cucchiara , Costantino Grana , Andrea Prati , Roberto Vezzani, Computer vision techniques for PDA accessibility of in-house video surveillance, First ACM SIGMM international workshop on Video surveillance, November 02-08, 2003, Berkeley, California [doi>10.1145/982452.982464]
• 4: R. Cucchiara, C. Grana, A. Prati, R. Vezzani, "Probabilistic Posture Classification for Human Behaviour Analysis" in press on IEEE SMC Transactions, Part A: Systems and Humans, special issue on Ambient Intelligence, 2004
• 5: R. Cucchiara, C. Grana, M. Piccardi, A. Prati, "Detecting Moving Objects, Ghosts and Shadows in Video Streams" in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, n. 10, pp. 1337--1342, 2003
• 6: Ismail Haritaoglu , Davis Harwood , Larry S. David, W4: Real-Time Surveillance of People and Their Activities, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.22 n.8, p.809-830, August 2000 [doi>10.1109/34.868683]
• 7: Michael Isard , Andrew Blake, A Smoothing Filter for CONDENSATION, Proceedings of the 5th European Conference on Computer Vision-Volume I, p.767-781, June 02-06, 1998
• 8: S. Khan, M. Shah, "Tracking People in Presence of Occlusion", Asian Conf. on Computer Vision, Taiwan, Jan 2000.
• 9: A. J. Lipton, et al. "Moving target classification and tracking from real-time video" IEEE Image Understanding Workshop, 1998, pp. 129--136.
• 10: Stephen J. McKenna , Sumer Jabri , Zoran Duric , Harry Wechsler, Tracking Interacting People, Proceedings of the Fourth IEEE International Conference on Automatic Face and Gesture Recognition 2000, p.348, March 26-30, 2000
• 11: H.T. Nguyen, and A. W.M. Smeulders. Template tracking using color invariant pixel features. In Proc. ICIP'02, Vol 1, pp. 569--573, Rochester, 2002.
• 12: P. Pérez, C. Hue, J. Vermaak and M. Gangnet. Color-based probabilistic tracking. ECCV'2002, Copenhagen, Denmark, June 2002
• 13: Multiple People Tracking Using an Appearance Model Based on Temporal Color, Proceedings of the International Conference on Pattern Recognition, p.4643, September 03-08, 2000
• 14: A. Senior, et al. "Tracking people with probabilistic appearance models", Int. Workshop on Perf. Eval. of Tracking and Surveillance Systems, 2002.
• 15: Nils T. Siebel , Stephen J. Maybank, Fusion of Multiple Tracking Algorithms for Robust People Tracking, Proceedings of the 7th European Conference on Computer Vision-Part IV, p.373-387, May 28-31, 2002
• 16: Chris Stauffer , W. Eric L. Grimson, Learning Patterns of Activity Using Real-Time Tracking, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.22 n.8, p.747-757, August 2000 [doi>10.1109/34.868677]
• 17: T. Zhao, R. Nevatia and F. Lv, Segmentation and Tracking of Multiple Humans in Complex Situations, CVPR Kauai, Hawaii, Dec., 2001.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.8 Scene Analysis
          Subjects: Motion

Additional Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.8 Scene Analysis
          Subjects: Tracking

General Terms:
Algorithms, Design, Experimentation, Performance, Theory

Keywords:
occlusions, people tracking, probabilistic models, video surveillance