全景视频双环带映射算法

发布时间：2018-08-25 08:21

【摘要】：针对全景视频映射过程中局部区域变形过大、冗余数据量极高的问题,提出了一种双环带映射算法(DRP)。首先,根据球面视频的几何特点,结合人眼视度(HVS)这一视觉特性,用两个相互正交的环形区域,将球面视频分割成14个大小相近的区域;然后根据空域采样定理,采用兰索斯插值法,将这14个区域对应的球面视频内容映射为14个大小相等的矩形视频;最后,根据最新视频编码标准的特点,重新排列这14个矩形视频,得到符合编码器标准的紧凑的全景视频。实验结果表明,与经纬图映射算法(ERP)、八面体映射算法(OHP)、二十面体映射算法(ISP)相比,DRP算法在视频压缩性能方面有良好的表现;其中同最流行的ERP算法相比,码率平均降低8.61%,明显提升了视频编码效率。
[Abstract]:In order to solve the problem that the local area deformation is too large and the redundant data is extremely high in panoramic video mapping process, a double loop band mapping algorithm (DRP). Is proposed. Firstly, according to the geometric characteristics of spherical video, combined with the visual characteristic of human visual degree (HVS), the spherical video is divided into 14 regions of similar size by using two orthogonal ring regions, and then according to the spatial sampling theorem, The spherical video content corresponding to these 14 regions is mapped to 14 rectangular video of equal size by Ransors interpolation. Finally, according to the characteristics of the latest video coding standards, the 14 rectangular video are rearranged. Get a compact panoramic video that meets the encoder standard. The experimental results show that compared with the longitude and latitude map mapping algorithm (ERP), octahedron mapping algorithm (OHP), icosahedron mapping algorithm (ISP) algorithm has a good performance in video compression performance, and compared with the most popular ERP algorithm, The average bit rate is reduced by 8.61, which obviously improves the efficiency of video coding.
【作者单位】：上海大学通信与信息工程学院;
【基金】：国家自然科学基金资助项目(61271212) 国家863计划项目(2015AA015903)~~
【分类号】：TN919.81

【相似文献】