Face image super-resolution method based on improved deep iterative collaborative network

Abstract

The invention discloses a face image super-resolution method based on an improved deep iterative collaborative network, and the method comprises the steps: 1), carrying out the early-stage data processing, and obtaining low-resolution face image data; 2) inputting a low-resolution face image into an image super-resolution sub-network, and processing the face image by a shallow feature extraction module to obtain shallow features; 3) the output of the shallow feature and prior information extraction sub-network in the last iteration process is sent to the image super-resolution sub-network to obtain high-resolution features, and a reconstruction module reconstructs the high-resolution features to obtain a high-resolution image; 4) inputting the high-resolution image into the prior information extraction sub-network, and outputting a face key point thermodynamic diagram, a semantic analysis diagram and intermediate layer features at the same time; and (5) repeating the steps (3) and (4), and carrying out iteration for N times to obtain final high-resolution image output. According to the method, the structure information of the face image is fully considered, the super-resolution effect of the face image is better, the parameter scale is smaller, and the time overhead is smaller.

Description

technical field [0001] The invention relates to the technical field of deep learning image processing, in particular to a face image super-resolution method based on an improved deep iterative collaboration network. Background technique [0002] Image super-resolution refers to the restoration of low-resolution images into high-resolution images, while face image super-resolution is a specific application of image super-resolution technology in the field of face. Many technologies related to the face field, such as face recognition or face beautification, will be extremely degraded when applied to low-resolution face images. However, in real scenes, due to the influence of shooting equipment, distance and noise, there are often only low-resolution face images. Therefore, face image super-resolution is crucial for these face domain-related technologies. [0003] As deep learning shines brightly in the image field, image super-resolution technology based on deep learning has...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the shortcomings of the existing deep iterative collaborative network, considering the problems of insufficient use of face structure information, large model parameter scale, space and time overhead, etc., and propose a face recognition algorithm based on an improved deep iterative collaborative network. Image super-resolution method, this method makes full use of face structure information to promote the image super-resolution process, through training, the network can be continuously optimized to improve the super-resolution effect of face images

Images