【Academic report】——Screen content coding

Title: Screen content coding

Time: Dec.7   10:00 a.m.

Location: 307B

Reporter: Sun Huifang

【Abstract】

Screen content video has evolved from a niche to a mainstream due to therapid advances in mobile and cloud technologies. Real-time, low-latency transport of screen visuals between devices in the form of screen content video is becoming prevalent in many applications, e.g. wireless display, screen mirroring, mobile or external display interfacing, screen/desktop virtualization and cloud gaming. Today's commonly-used video coding methods, however, have been developed primarily with camera-captured content in mind. These new applications create an urgent need for efficient coding of screen content video, especially as the support of 4k or even 8k resolution begins to achieve mass market appeal. Screen content video coding poses numerous challenges. Such content usually features a mix of computer generated graphics, text, and camera-captured images/video. With their distinct signal characteristics, content adaptive coding becomes necessary. This is without mentioning the varied level of the human's visual sensitivity to distortion in different types of content; visually or mathematically lossless quality may be required for all or part of thevideo. In this talk, Prof. Sun will introduce the progress of screen content coding.