Enhancing change detection in low-quality surveillance footage using markov random fields

ABSTRACT

Urban surveillance footage can be of poor quality, partly due to the low quality of the camera and partly due to harsh lighting and heavily reflective scenes. For some computer surveillance tasks very simple change detection is adequate, but sometimes a more detailed change detection mask is desirable, eg, for accurately tracking identity when faced with multiple interacting individuals and in pose-based behaviour recognition. We present a novel technique for enhancing a low-quality change detection into a better segmentation using an image combing estimator in an MRF based model.

REFERENCES

• 1: Y. Boykov and M.-P. Jolly. Interactive graph cuts for optimal boundary and region segmentation of objects in N-D images. In Proc ICCV, 2000.
• 2: Yuri Boykov , Vladimir Kolmogorov, An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.26 n.9, p.1124-1137, September 2004 [doi>10.1109/TPAMI.2004.60]
• 3: G. Celeux, J. S. Marques, and J. Nascimento. Learning switching dynamic models for objects tracking. Pattern Recognition, 37(9):1835--1840, 2004.
• 4: D. Chetverikov. CVOnline: Hausdorff distance. homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/CHETVERIKOV/mysquash/node4.html.
• 5: Olivier Devillers , Mordecai J. Golin, Incremental algorithms for finding the convex hulls of circles and the lower envelopes of parabolas, Information Processing Letters, v.56 n.3, p.157-164, Nov. 10, 1995 [doi>10.1016/0020-0190(95)00132-V]
• 6: Ahmed Elgammal , Ramani Duraiswami , Larry S. Davis, Efficient Kernel Density Estimation Using the Fast Gauss Transform with Applications to Color Modeling and Tracking, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.25 n.11, p.1499-1504, November 2003 [doi>10.1109/TPAMI.2003.1240123]
• 7: M. Evans and J. Ferryman. Group stochastic search for object detection and tracking. In IEEE Int Conf on Advanced Video and Signal-Based Surveillance, 2005.
• 8: P. F. Felzenswalb and D. P. Huttenlocher. Efficient belief propagation for early vision. In Proc CVPR, 2004.
• 9: Daniel Freedman , Matthew W. Turek, Illumination-Invariant Tracking via Graph Cuts, Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2, p.10-17, June 20-26, 2005 [doi>10.1109/CVPR.2005.182]
• 10: Dariu Gavrila, Pedestrian Detection from a Moving Vehicle, Proceedings of the 6th European Conference on Computer Vision-Part II, p.37-49, June 26-July 01, 2000
• 11: D. Greig, B. Porteous, and A. Seheult. Exact maximum a posteriori estimation for binary images. J Royal Statistical Soc, Series B, 51(2):271--279, 1989.
• 12: M. Hassaballah , Saleh Omran , Youssef B. Mahdy, A Review of SIMD Multimedia Extensions and their Usage in Scientific and Engineering Applications, The Computer Journal, v.51 n.6, p.630-649, November 2008 [doi>10.1093/comjnl/bxm099]
• 13: V. Kolmogorov. Max-flow MRF software. www.adastral.ucl.ac.uk/~vladkolm/software.html.
• 14: M. Latzel and J. K. Tsotsos. A robust motion detection and estimation filter for video signals. In Proc ICIP, 2001.
• 15: C.-S. Lee and A. Elgammal. Modeling view and posture manifolds for tracking. In Proc ICCV, 2007.
• 16: Z. Lin, L. S. Davis, D. Doermann, and D. DeMenthon. Hierarchical part-template matching for human detection and segmentation. In Proc ICCV, 2007.
• 17: MuHAVi-MAS dataset. http://dipersec.king.ac.uk/MuHAVi-MAS.
• 18: H. Ragheb, S. Velastin, P. Remagnino, and T. Ellis. Human action recognition using robust power spectrum features. In IEEE ICIP, 2008.
• 19: V. Sharma and J. W. Davis. Simultaneous detection and segmentation of pedestrians. In Proc ICCV, 2007.
• 20: Jonathan Starck , Adrian Hilton, Surface Capture for Performance-Based Animation, IEEE Computer Graphics and Applications, v.27 n.3, p.21-31, May 2007 [doi>10.1109/MCG.2007.68]
• 21: S. Vicente, V. Kolmogorov, and C. Rother. Graph cut based image segmentation with connectivity priors. In Proc CVPR, 2008.
• 22: Y. Zeng, D. Samaras, W. Chen, and Q. Peng. Topology cuts: A novel Min-Cut/Max-Flow algorithm for topology preserving segmentation in N-D images. Technical report, Computer Science, Stony Brook University, 2007.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE

General Terms:
Algorithms

Keywords:
change detection, silhouette, surveillance