基于视觉显著性的立体视频编码分析-信息与通信工程专业论文.docxVIP

ABSTRACT Stereoscopic video is a collection of multiple video sequences, which capturing the same scene via two-view or multi-view camera. Compared with the single-channel video, stereoscopic video increases the depth information of the scene, and enhances the three-dimensional sense and realism of the video. Stereoscopic video has not widely used in practical applications yet since it has a large amount of data. Therefore, the efficient compression of the video data is significant for reducing the need for storage and transportation. In order to achieve a certain degree of compression, the prediction and compensation methods for intra-frame, inter-frame and inter-view are employed to exploit the temporal redundancy, spatial redundancy and inter-view redundancy, which are abundantly existed in stereoscopic video. However, the conventional stereoscopic video coding approaches ignore the diversity that human vision system (HVS) perceives on video scene. This thesis starts from the visual characteristics of the human eyes and studies stereoscopic video coding methods based on the visual saliency. A large number of physiological and psychological researches based on HVS have shown that while watching and comprehending video images, human will focus attention on several regions which may cause visual saliency. These regions can excellently represent the scene information and image content, they are worth concerning and are called region of interest (ROI). According to the different interesting degree and different sensitivity to distortion of different regions, depth information is adopted in this article to divide the video sequences to multi-level ROI, and appropriate measures are taken to adjust quantization parameter (QP) so that more bits are allocated for ROI to improve the video quality, and fewer bits for non-ROI to achieve more efficient compression. Multilevel region of interest guided bit allocation method can enhance the smoothness of ROI boundary, degrade b