Mode Information Guided CNN for Quality Enhancement of Screen Content Coding

Video quality enhancement methods are of great significance in reducing the artifacts decoded videos High Efficiency Coding (HEVC). However, existing mainly focus on improving natural sequences, not for screen content sequences that have drawn more attention than ever due to demands remote desktops and online meetings. Different from encoded by HEVC, Screen Content (SCC), an extension tool HEVC. Therefore, we propose a Mode Information guided CNN (MICNN) further improve at decoder side. To exploit characteristics extract mode information bitstream as input MICNN. Furthermore, limited number establish large-scale dataset train validate our Experimental results show proposed MICNN can achieve 3.41% BD-rate saving average. In addition, method consumes acceptable computational time compared with other video methods.