A flexible and uniform string matching technique for general screen content coding

Abstract

This paper proposes a flexible and uniform string matching technique named universal string matching (USM) for general screen content coding (SCC). USM uses two reference buffers for string matching: primary reference buffer (PRB) and secondary reference buffer (SRB), and includes three modes: general string (GS) mode, constrained string 1 (CS1) mode, and constrained string 2 (CS2) mode. PRB is used in GS mode and CS1 mode and SRB is used in GS mode and CS2 mode. Each of the three modes plays an essential role in SCC due to the diversity and comprehensiveness of the screen content. The experiments use HEVC SCC common test condition (CTC) for lossy coding. Compared with HEVC HM-16.6 + SCM-5.2 reference software of full frame search range for IBC and with ACT off, USM achieves an average Y BD-rate of −25.5% for four TGM (text and graphics with motion) test sequences from the SCC verification test suite and −5.5% for eight TGM test sequences from the HEVC SCC CTC test suite in all intra configuration with a small increase of encoding runtime and a small decrease of decoding runtime.