Content-Aware Video Seam Carving Based on Bag of Visual Cubes

Carving in consecutive frames with different criteria would usually result in discontinuity of visual perception. Therefore, how to preserve the visual continuity in video frames is the most critical issue in the field of video seam carving. In this paper, we propose a novel approach for modelling dynamic visual attention based on spatio temporal analysis in order to detect the focus of interest automatically. The continuously varied co-sited blocks in a video cube are first detected and their variations are characterized as a bag of visual cubes, which are further employed to determine a proper extent of salient regions in video frames. Once the proper extent through video cubes is determined, the carving process then can be conducted to find the global optimum. Our experiment shows that the proposed content-aware video seam carving based on spatio temporal bag of visual cubes can effectively generate resized videos while keeping their isotropic manipulation and the continuous dynamics of visual perception.