首页> 外文会议>International Conference on Signal Processing and Multimedia Applications >Data reuse in two-level hierarchical motion estimation for high resolution video coding
【24h】

Data reuse in two-level hierarchical motion estimation for high resolution video coding

机译:高分辨率视频编码的两级分层运动估计中的数据重用

获取原文

摘要

In hardware implementation of a video coding system memory access becomes a critical issue and the motion estimation (ME) module is the one which consumes most of the data access. To meet the requirements of HD specification, conventional data reuse schemes are not sufficient. In this paper, the hierarchical approach to ME is combined with the Level C and D data reuse schemes. Proposed two-level hierarchical ME algorithm reduces the external memory bandwidth by 77% and the on-chip memory size by 93% with reference to the Level C scheme, and computational complexity by over 99% with reference to the one-level full search (OLFS), achieving the results close to OLFS.
机译:在视频编码系统的硬件实现中,存储器访问成为一个关键问题,并且运动估计(ME)模块是消耗大多数数据访问的问题。为了满足高清规范的要求,传统的数据重用方案不足。在本文中,将ME的分层方法与C级和D数据重用方案相结合。提出的两级分层ME算法将外部内存带宽和片上存储器大小减少了93%,参考C级别方案,并参考单级全面搜索超过99%的计算复杂度( OLFS),实现靠近OLF的结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号