Perceptually Quasi-Lossless Compression of Screen Content Data Via Visibility Modeling and Deep Forecasting

Abstract

Screen content data, such as computer-generated photographs, desktop sharing, remote education, video game streaming and screenshot, is one of the most popular visual information carriers in Internet of Video Things. Although lossless compression can guarantee high quality of service for these screen content based industrial applications, it also causes considerable storage space and transmission bandwidth issues. To alleviate these challenges, in this article, we present a visually quasi-lossless coding approach to control the compression distortion below <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">visibility threshold</i> in the human visual system. Specifically, to better quantify the visual redundancy for screen content data, a new <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">visibility threshold</i> method is designed by incorporating blur sensitivity and oblique correction effects. Then, an end-to-end mapping between the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">visibility threshold</i> and quality control factor is learned and represented as a deep convolutional neural network. The experimental results demonstrate that the proposed method saves the average encoding bits up to 23.15% compared with the latest scheme under the same perceptual quality.