Method for reconstructing distributed video coding based on constraints on temporal-spatial correlation of video

Abstract

The invention discloses a method for reconstructing distributed video coding based on constraints on temporal-spatial correlation of video, which belongs to the technical field of video signal processing, and comprises the following steps: after completing the decoding of a Wyner-Ziv frame code stream, determining the pixel value or conversion coefficient value of the Wyner-Ziv frame into the range of [BL, BU] by a decoder; according to the information such as the statistic correlation between the Wyner-Ziv frame and the side information, the gradient information of the Wyner-Ziv frame in different directions, and grounds between adjacent frames, and the like, modeling the reconstruction of the Wyner-Ziv frame so as to solve the optimal solution problem of Markov random field; introducing the constraint on spatial correlation between adjacent pixels or conversion coefficients in sub-bands by defining energy functions, and calculating some parameters of the defined energy functions by analyzing and using the correlation between adjacent frames by grounds. In the invention, through introducing the constraints on temporal correlation and spatial correlation of the video signals, the subjective quality and objective quality of a reconstruction result can be simultaneously improved.

Description

technical field [0001] The present invention relates to a method in the technical field of video signal processing, in particular to a distributed video coding reconstruction method based on video spatiotemporal correlation constraints. Background technique [0002] Video codec technology is crucial for applications such as video storage and transmission. Existing video coding technologies (such as H.264 / AVC, MPEG-x series standards) perform motion estimation and motion compensation at the coding end to eliminate temporal redundancy of video signals. Although this scheme can obtain better rate-distortion performance, it requires the computational complexity of the encoder to be much higher than that of the decoder (usually 5-10 times). Distributed video coding (also known as Wyner-Ziv video coding) technology is a new video coding framework, which meets the requirements of some new applications that have appeared recently, such as mobile phone camera, wireless multimedia se...

AI Technical Summary

Problems solved by technology

These schemes model the pixel values ​​(or transform coefficient values) of video frames as independent and identically distributed random sources, and do not consider the correlation constraints between adjacent pixels (transform coefficients in the same subband)

Images