Face super-resolution method based on dense residual attention face prior network

Abstract

The invention discloses a face super-resolution method based on a dense residual attention face prior network. The method comprises the following steps: respectively constructing a dense residual attention module, a face structure prior prediction module and an up-sampling module which are connected by a jumper; connecting a dense residual attention module and a facial structure prior prediction module which are connected by a jumper in parallel, and then cascading an up-sampling module and an image reconstruction layer to construct a dense residual attention facial prior network; preprocessing the published data set, and dividing the processed data into a training set and a test set; training a dense residual attention face prior network; and inputting the images in the test set into thetrained dense residual attention face prior network, and outputting the reconstructed high-resolution face image. According to the method, the dense residual attention face prior network is trained, so that super-resolution processing of the tested face image is realized, high-frequency details of the face can be effectively recovered, and meanwhile, the super-resolution face image with identity information is reserved.

Description

technical field [0001] The invention belongs to the technical field of image processing and human face super-resolution, and in particular relates to a human face super-resolution method based on dense residual attention facial prior network. Background technique [0002] In face recognition tasks, it is often expected to obtain high-quality images with high resolution, clear and noise-free. This is because high-quality face images not only have better visual effects, but also contain a lot of detailed information required in subsequent processing. However, in the actual acquisition and transmission system, due to the limitations of the acquisition environment, imaging hardware, and network bandwidth, the resolution of the acquired face images is often relatively small, which greatly affects the accuracy of subsequent face recognition tasks. Improving the hardware conditions of the imaging system and controlling the acquisition environment are the most direct ways to improv...

AI Technical Summary

Problems solved by technology

[0004] The existing face super-resolution methods usually fail to make full use of the correlation between facial prior information and non-local information, resulting in the lack of high-frequency information of the face and generating high-scoring face images with too many artificial artifacts

At the same time, most face super-resolution algorithms only consider the mean square error loss when designing. Although this loss can obtain better objective indicators, it fails to take into account the inherent information of the face.

Images