Video based behavior detection algorithm

Abstract Action recognition based on video surveillance becomes possible because of the rapid development action recognition, temporal and spatial-temporal technology. A video-based behavior detection algorithm designed to find information interest from videos. In process detection, feature extraction is often carried out space time dimensions. However, calculation amount videos sent into deep convolutional network much higher than that pictures. Therefore, design lightweight conducive reducing training costs, realizing efficient calculation, enabling researchers quickly apply it practical scenarios.