Efficient video encoding method based on vision attention priority

Abstract

The invention discloses an efficient video encoding method based on a vision attention priority. The method is centered on two contents including a vision attention priority annotation based on a movement-texture feature and a high-quality encoding optimization based on the vision attention priority. The method comprises the following steps: firstly, performing the attention annotation and priority judgment on an encoding unit when the movement grading of an encoding object is acted as a leading vision attention feature and texture information of the encoding object is acted as an auxiliary vision attention feature; and secondly, performing self-adaptive encoding conversion coefficient regulation to realize difference encoding aiming at regions in different attentions, and guaranteeing more optimal subjective vision quality under the same code rate.

Description

technical field [0001] The invention belongs to the technical field of video processing, and relates to a high-efficiency video coding method based on visual attention priority. Background technique [0002] Today, with the deepening of technological globalization, the accelerated popularization of video information has triggered the explosive growth of video data, and at the same time promoted the continuous improvement of video coding technology marked by high efficiency. In this context, video coding technology based on the characteristics of human visual perception came into being. [0003] The Human Visual System (HVS) can fully perceive various visual information such as brightness, color, motion, and texture in video scenes, and is one of the important means for the human eye to acquire and process external information. At the same time, since the human eye is the main receiver of video data, a proper combination of the human visual system and video coding technology...

AI Technical Summary

Problems solved by technology

However, the existing international video coding standards, such as H.265 / HEVC, do not distinguish the visual significance of different areas or objects in the image when compressing video data. Therefore, in bit rate allocation or In the process of computing resource allocation, uniform allocation standards are adopted, and there is no consideration of "different from person to person"

Images