CUDA-based H.264/AVC Deblocking Filtering

摘要

In H.264/AVC decoder,adaptive deblocking filter is one of the most complex modules and it takes 33% of the whole decoding process time. Its optimization, not only in algorithm level but also in code level, is a very impor tant topic. In this paper, CUDA technology is proposed for speeduping the deblocking process. First, a frame is divided into 16×16 macroblocks and/or 4×4 subblocks. Second, edges of each block are highly parallelly filtered by SIMT(single instruction multiple thread) CUDA in structions. Comparison of the time consumption by pro cessing a 640×480(4:2:0) video on CUDA and CPU is carried out. The proposed CUDA approach offers al most 15x improvement on the decoding efficiency.