Lossless frame memory recompression for video codec preserving random accessibility of coding unit

Abstract

In recent video applications such as MPEG or H.264/AVC, the bandwidth requirement for frame memory has become one of the most critical problems. Compressing pixel data before storing in off-chip frame memory is required to alleviate this problem. This thesis proposed a lossless frame memory recompression scheme including 1) a lossless pixel compression algorithm with hierarchical prediction scheme and entropy coding scheme or low latency decoding process which maintains the balance between memory access cycles and decompression cycles, 2) an efficient organization of the address table with pattern based compression method to preserve random accessibility for the compressed data with the small size of embedded memory, and 3) frame memory placement scheme for the compressed data to reduce the effective access time of SDRAM by suppressing row switching. Experimental results show that the proposed method reduces the frame data to 46% compared to that of the uncompressed one with H.264/AVC high profile encoder system, where 6.1kB of SRAM is required for the address table of full HD video.