Pedestrian Detection Using Global-Local Motion Patterns
We propose a novel learning strategy called Global-Local Motion Pattern Classification (GLMPC) to localize pedestrian-like motion patterns in videos. Instead of modeling such patterns as a single class that alone can lead to high intra-class variability, three meaningful partitions are considered - left, right and frontal motion. An AdaBoost classifier based on the most discriminative eigenflow weak classifiers is learnt for each of these subsets separately. Furthermore, a linear three-class SVM classifier is trained to estimate the global motion direction. To detect pedestrians in a given image sequence, the candidate optical flow sub-windows are tested by estimating the global motion direction followed by feeding to the matched AdaBoost classifier. The comparison with two baseline algorithms including the degenerate case of a single motion class shows an improvement of 37% in false positive rate.
KeywordsSupport Vector Machine Motion Pattern Global Motion Frontal Motion Linear Support Vector Machine
Unable to display preview. Download preview PDF.
- 1.Oren, M., Papageorgiou, C., Sinha, P., Osuna, E., Poggio, T.: Pedestrian detection using wavelet templates. CVPR, 193–199 (1997)Google Scholar
- 2.Dalal, N., Triggs, B.: Histograms of Oriented Gradients for Human Detection. CVPR 1, 886–893 (2005)Google Scholar
- 3.Viola, P., Jones, M., Snow, D.: Detecting Pedestrians Using Patterns of Motion and Appearance. ICCV 2, 734–741 (2003)Google Scholar
- 4.Fablet, R., Black, M.J.: Automatic Detection and Tracking of Human Motion with a View-Based Representation. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2352, pp. 476–491. Springer, Heidelberg (2002)Google Scholar
- 5.Sidenbladh, H.: Detecting Human Motion with Support Vector Machines. ICPR 2, 188–191 (2004)Google Scholar
- 6.Goel, D., Chen, T.: Real-time Pedestrian Detection using Eigenflow. In: IEEE International Conference on Image Processing, IEEE Computer Society Press, Los Alamitos (2007)Google Scholar
- 11.Viola, P., Jones, M.: Rapid Object Detection using a Boosted Cascade of Simple Features. CVPR (2001)Google Scholar