Multi-pose pedestrian detection based on posterior HOG feature

Pedestrian detection remains one of the challenging tasks in the area of computer vision. A multi-pose pedestrian detection method based on posterior HOG feature is proposed. Firstly, the generality information of gradient feature energy is computed with all pedestrian samples. The posterior HOG feautre is obtained by weighting the HOG feature of individual pedestrian sample with the computed gradient feature energy. The posterior HOG feature can capture the contours and edges of pedesrtians, and significantly reduce the influence of complex and cluttered background. Secondly, pedestrians of different poses and views are divided into subclasses with S-Isomap and K-means algorithm. A classifier is trained for each subclass. Finally, a multi-pose-view ensemble classifier is trained to combine the output values of different subclass classifiers with an equally weighted sum rule. Experimental results on different datasets suggest that the proposed posterior feature outperforms the classic HOG feature and other typical features. Compared with the existing methods, by combining the posterior feature and the multi-pose-view ensemble classifier, the proposed method boosts the detection accuracy effectively. ©, 2015, Chinese Institute of Electronics. All right reserved.