Online video denoising system with advanced circulation module

Abstract

The invention discloses an online video denoising system with an advanced circulation module, and belongs to the technical field of video denoising. The invention aims to solve the problem that an existing video denoising system based on a bidirectional recurrent neural network cannot perform online processing. Comprising a forward circulation module which is used for processing a current frame noise image, an adjacent previous frame noise image and forward features of the adjacent previous frame noise image to obtain the forward features of the current frame noise image; the advanced circulation module is used for processing the advanced features of the adjacent selected future frame, the noise image of the previous frame of the adjacent selected future frame and the noise image of the previous frame of the adjacent selected future frame to obtain the advanced features of the noise image of the adjacent selected future frame; the alignment module is used for aligning the advanced features of the future frame from the noise image of the future frame to the noise image of the current frame to obtain the aligned advanced features; and the decoding module is used for fusing the forward features of the current frame and the aligned advanced features to obtain a denoised image of the current frame. The method is used for online video denoising.

Description

technical field [0001] The invention relates to an online video denoising system with an advanced loop module, and belongs to the technical field of video denoising. Background technique [0002] Video denoising aims to restore clear video from noisy video. The main cause of noise in video is poor recording equipment or low-light environment. With the development and popularity of video applications (such as live broadcast, online conferences, etc.), obtaining high-quality video content through video denoising has become a research hotspot. [0003] In recent years, image denoising methods based on neural networks have achieved good results, but different network structures still have their own shortcomings. For example, convolutional neural networks cannot utilize long-distance timing information; unidirectional recurrent neural networks cannot utilize future timing information; bidirectional recurrent neural networks cannot be processed online and cannot be used for onlin...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the existing video denoising system based on bidirectional cyclic neural network cannot be processed online, the present invention provides an online video denoising system with an advanced cyclic module

Images