Feature-matching based motion prediction for high efficiency video coding in cloud
Feature-matching based motion prediction for high efficiency video coding in cloud
复制标题
DOI：
10.1109/icmew.2015.7169778
复制
发表时间：
2015-07
期刊：
影响因子：
--
通讯作者：
Xianguo Zhang;Shiqi Wang;Shanshe Wang;Siwei Ma;Wen Gao
中科院分区：
文献类型：
--
作者：
Xianguo Zhang;Shiqi Wang;Shanshe Wang;Siwei Ma;Wen Gao
图像和视频帧的视觉特征在诸如计算机视觉和视觉搜索的广泛研究领域中已经变得普遍和快速发展。对于实时检索应用，紧凑的视觉特征应该在云中传输并存储在服务器端。这些局部特征描述符的特点是对相机运动、光照变化、遮挡和不同视点引起的方差具有不变性。受这些性质的启发，在这项工作中，典型的尺度不变特征变换（SIFT）描述符被用来提高视频编码效率。特别地，使用SIFT匹配的预测运动用于高效率视频编码（HEVC）标准中的合并模式和运动矢量预测（MVP）。进一步提出了一种分层运动推导框架，旨在实现鲁棒和有效的MVP。实验结果表明，该方法能有效地提高图像的编码性能。
Visual features of images and video frames have become pervasive and maturely developed in extensive research fields such as computer vision and visual search. For realtime retrieval applications, the compact visual features should be transmitted and stored at server side in cloud. These local feature descriptors are characterized by the invariance properties for the variances caused by camera motion, illumination changing, occlusion and different viewpoints. Inspired by these properties, the typical scale-invariant feature transform (SIFT) descriptor is leveraged to improve the video coding efficiency in this work. In particular the predicted motion using SIFT matching is used for merge mode and motion vector prediction (MVP) in the high efficiency video coding (HEVC) standard. A hierarchical motion derivation framework aiming at achieving robust and effective MVP is further proposed. Experimental results have shown that the proposed method can efficiently improve the coding performance according to the accurate feature-matching.