Accurate Human Detection by Appearance and Motion

Abstract

SUMMARYIn this paper, a human detection method is developed. An appearance based detector and a motion based detector are proposed respectively. A multi scale block histogram of template feature (MB-HOT) is used to detect human by the appearance. It integrates the gray value information and the gradient value information, and represents the relationship of three blocks. Experiment on INRIA dataset shows that this feature is more discriminative than other features, such as histogram of orientation gradient (HOG). A motion based feature is also proposed to capture the relative motion of human body. This feature is calculated in optical flow domain and experimental result in our dataset shows that this feature outperforms other motion based features. The detection responses obtained by two features are combined to reduce the false detection. Graphic process unit (GPU) based implementation is proposed to accelerate the calculation of two features, and make it suitable for real time applications. key words: human detection, multi scale block histogram of template, Graphics process unit

Implementation of Scale and Rotation Invariant On-Line Object Tracking Based on CUDA

“…A serial processing problem can often be partitioned into coarse sub problems solved independently in parallel. Now GPU parallel computing has seen increasing applications in the domain of image processing and computer vision, such as face detection [3], feature matching [4], human detection [5] tern classification [6]. Despite the tremendous computing capacity of GPU, utilizing it to deal with practical problems is non-trivial.…”