Human Detection Based on Optical Flow and Spare Geometric Flow

Different from most of the previous work, we detect motion human by region segmentation and classification through machine learning. In our method, based on optical flow, region segmentation is carried firstly and then, based on geometric flow, Bandelet transform is used to do feature extraction and classification. Some treatments were carried after optical flow field computation to denoising and some improvements in Bandelet transform were used to reduce time cost of feature extraction. The results of motion human detection experiments indicate that the proposed method can segment motion region more clearly and improve the performance of classifier effectively. It can be used to do real-time motion human detection in videos.