Video coding method and device and electronic equipment

Abstract

The invention provides a video coding method and device and electronic equipment. The video coding method comprises the following steps: extracting a video frame sequence with a preset frame number from a video to be coded; inputting the video frame sequence into a pre-trained video processing model to obtain a processed video frame sequence; wherein the video processing model is used for carrying out de-noising processing on the video frame sequence, and the de-noising processing process comprises the step of carrying out down-sampling in a mode of rearranging pixel points of a video frame image; and performing coding compression processing on the processed video frame sequence to obtain a compressed video code stream. According to the embodiment of the invention, the de-noising network model is pre-trained, and the video code stream is pre-processed by using the de-noising network model before the video code stream is coded, so that the coding rate bandwidth consumed by noise signals can be saved; moreover, the de-noising network model performs down-sampling in a mode of rearranging the pixel points of the video frame image, thereby realizing lossless image information in the down-sampling process, and improving the quality of the decoded video.

Description

technical field [0001] The present invention relates to the technical field of video processing, in particular to a video encoding method, device and electronic equipment. Background technique [0002] The dissemination of video information often requires a certain link process, such as: video acquisition-encoding-storage distribution-decoding-playback process, wherein, in the encoding process, part of the encoded bit stream is consumed by noise signals, and the noise signal itself It cannot provide more information to the picture, but it will destroy the coherence and semantic integrity of the picture, and waste the encoding bit rate bandwidth. [0003] At present, preprocessing methods for video images often focus on reducing noise in video images, so as to achieve the purpose of bit rate saving. Among them, image noise can be divided into many types, such as Gaussian white noise, compression noise (such as ringing, block noise) and so on. In real natural images, the typ...

AI Technical Summary

Problems solved by technology

In real natural images, the types of noise are very complex. Traditional image processing and denoising algorithms are often only aimed at a certain type of noise, and the conditions of use are ideal. The denoising method is single and the denoising effect is not good. The noise signal in consumes more coding bit rate

[0004] Generally speaking, the existing video coding method, because the denoising effect of the preprocessing of the video image before coding is poor, resulting in a lot of coding bit rate bandwidth wasted by the noise signal when the video is coded later.

Images