Efficient algorithm for all-hardware motion prediction in H.264/AVC video encoding

H.264/AVC视频编码中全硬件运动预测的高效算法

• 批准号: 412539-2011
• 负责人: Wang, Zhou
• 金额: $ 1.82万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2011
• 资助国家: 加拿大
• 起止时间: 2011-01-01 至 2012-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=491095
• 关键词: Efficient; algorithm; hardware; motion; prediction

Recent years have witnessed a dramatically increased demand for digital video services in a broad range of network communication environments. Due to technology constraints, current low-cost Internet video services, such as those provided by Youtube and Skype, only allow for low-resolution and low-quality video transmission. High-performance, low cost, and real-time high definition (HD) digital video encoders are highly desirable that facilitate high quality video broadcasting, videoconferencing and video-on-demand services. The system-on-chip video codec product by System-On-Chip (SOC) Technologies provides one of the most promising solutions in the market based on its single-clock driven, all-hardware architecture. The objective of the current NSERC Engage project is to improve the motion estimation engine in SOC's video encoder product, so as to further increase the compression ratio and thus reduce the required bandwidth for real-time

近年来，在广泛的网络通信环境中，对数字视频服务的需求急剧增加。由于技术的限制，目前低成本的互联网视频服务，如Youtube和Skype提供的，只允许低分辨率和低质量的视频传输。高性能、低成本、实时高清晰度（HD）数字视频编码器是实现高质量视频广播、视频会议和视频点播服务的迫切需要。片上系统（SOC）技术公司的片上系统视频编解码器产品基于其单时钟驱动的全硬件架构，提供了市场上最有前途的解决方案之一。当前NSERC Engage项目的目标是改进SOC视频编码器产品中的运动估计引擎，从而进一步提高压缩比，从而减少实时所需的带宽