Three-dimensional video coding method and device

Abstract

The invention relates to a three-dimensional video coding method and a three-dimensional video coding device. The three-dimensional video coding method comprises the steps of: acquiring a virtual drawing block obtained through virtual drawing by a B video block and a corresponding coded or uncoded A video block, or a virtual viewpoint image block corresponding to the B video block, and regarding the virtual drawing block or the virtual viewpoint image block as a reference block; coding the B video block in a current coding mode to obtain a precoding B video block, and acquiring a reconstructed virtual drawing block obtained through virtual drawing by the precoding B video block and the corresponding coded or uncoded A video block; calculating space domain distortions and time domain distortions of the reference block and the reconstructed virtual drawing block, merging the space domain distortions and the time domain distortions to obtain a drawing distortion; loading a Lagrangian multiplier of a B video frame to obtain a precoding bit number of the B video block, and calculating a rate-distortion cost according to the drawing distortion, the Lagrangian multiplier and the precoding bit number; traversing codes of all coding modes, regarding the coding mode with the minimal rate-distortion cost as the optimal coding mode of the B video block; and acquiring a code of a next B video block until coding of the B video frames to be coded is completed. Therefore, the three-dimensional video coding efficiency is improved.

Description

technical field [0001] The present invention relates to the field of video coding, in particular to a three-dimensional video coding method and device. Background technique [0002] With the rapid development of digital media technology and the continuous improvement of living standards, people put forward higher requirements for video viewing, and the traditional flat (two-dimensional) video viewed with a fixed viewpoint can no longer meet people's needs. 3D TV and free-viewpoint video emerged as the times require. 3D TV and free-viewpoint video increase the depth information of scene objects, so that the image is no longer limited to the screen plane, as if it can go out of the screen, allowing users to feel the sense of depth and immersion. The 3D scene representation method of Multi-view Video plus Depth (MVD) has gradually become the mainstream data representation format in multimedia systems. MVD video data includes multi-channel color and depth video, and the amount o...

AI Technical Summary

Problems solved by technology

In the existing coding, the rate-distortion optimization only considers the distortion of the transmitted viewpoint image, but does not consider the distortion of the virtual viewpoint synthesized, and the subjective and objective quality of the 3D video cannot be optimal.

Images