Enhanced pipelined architecture of H.264/AVC intra prediction

作者：

Highlights：

• The pipelined method in and between macro blocks accelerates the encoding speed.

• All 4×4 mode formula sharing method allows 16-pixel and 9-mode parallelism.

• Streamlined reconstruction method optimizes the computational redundancy.

• CAVLC encoder with three parallel units improves entropy coding speed signiﬁcantly.

摘要

•The pipelined method in and between macro blocks accelerates the encoding speed.•All 4×4 mode formula sharing method allows 16-pixel and 9-mode parallelism.•Streamlined reconstruction method optimizes the computational redundancy.•CAVLC encoder with three parallel units improves entropy coding speed signiﬁcantly.

论文关键词：H.264/AVC,Intra prediction,Pipeline,VLSI architecture

论文评审过程：Received 14 March 2015, Revised 27 November 2015, Accepted 30 November 2015, Available online 12 December 2015, Version of Record 29 December 2015.

论文官网地址：https://doi.org/10.1016/j.image.2015.11.009