- Pedestrian detection
Pedestrian detection is an essential and significant task in any intelligent
video surveillance system, as it provides the fundamental information forsemantic understanding of thevideo footages.Challenges for pedestrian detection
* Various style of clothing in appearance
* Different possible articulations
* The presence of occluding accessories
* Frequent occlusion between pedestriansExisting Approaches
Despite the challenges, pedestrian detection still remains an active research area in
computer vision in recent years. Numerous approaches have been proposed.Holistic Detection
Detectors are trained to search for pedestrians in the video frame by scanning the whole frame. The detector would “fire” if the image features inside the local search window meet certain criteria. Some methods employ global features such as edge template [ N. Dalai, B. Triggs, I. Rhone-Alps, and F. Montbonnot. “Histograms of oriented gradients for human detection”, "IEEE Computer Society Conference on Computer Vision and Pattern Recognition" (CVPR), page 1:886-893, 2005 ] , others uses local features like histogram of oriented gradient descriptors
Histogram of oriented gradients [ C. Papageorgiou and T. Poggio, “A Trainable Pedestrian Detection system”, "International Journal of Computer Vision" (IJCV), page 1:15-33,2000 ] . The drawback of this approach is that, the performance can be easily affected by background clutter and occlusions.Part-based Detection
Pedestrians are modeled as collections of parts. Part hypotheses are firstly generated by learning local features, which includes edgelet features [Bo Wu and Ram Nevatia, “Detection of Multiple, Partially Occluded Humans in a Single Image by Bayesian Combination of Edgelet Part Detectors”, "IEEE International Conference on Computer Vision" (ICCV), pages 1:90-97, 2005] the orientation features [ Mikolajczyk, K. and Schmid, C. and Zisserman, A. “Human detection based on a probabilistic assembly of robust part detectors”, "The European Conference on Computer Vision" (ECCV), volume 3021/2004, pages 69-82, 2005 ] , and etc. These part hypotheses are then joined to form the best assembly of existing pedestrian hypotheses. Though this approach is attractive, part detection itself is a difficult task.
Patch-based Detection
Recently Bastian et al. [ B.Leibe, E. Seemann, and B. Schiele. “Pedestrian detection in crowded scenes” "IEEE Conference on Computer Vision and Pattern Recognition"(CVPR), pages 1:878-885, 2005] proposed an approach combining both the detection and segmentation with the name Implicit Shape Model (ISM). A codebook of local appearance is learned during the training process. In the detecting process, extracted local features are used to match against the codebook entries, and each match casts one vote for the pedestrian hypotheses. Final detection results can be obtained by further refining those hypotheses. The advantage of this approach is only a small number of training images are required.
References
Wikimedia Foundation. 2010.
