Model training, image processing method, device, electronic device and storage medium

Abstract

The present invention provides a model training, image processing method, device, electronic equipment, and computer-readable storage medium. The training method includes: obtaining a training sample set, and the training sample set includes a first image and a third image with clear human face regions , and the second image after adding noise to the face area of ​​the first image, the face area in the third image and the first image belong to the same user; obtain the first face semantic segmentation result of the second image and the third image The second human face semantic segmentation result; the second image and the first human face semantic segmentation result, and the third image and the second human face semantic segmentation result are input to the neural network model to obtain the fourth image; identify the fourth image and The difference data on the face area between the first images; the neural network model is iteratively updated according to the difference data; the iteratively updated neural network model is used to reconstruct the face area of ​​the image with the blurred face area to generate a face A clear image of the area.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular to a model training, image processing method, device, electronic equipment and storage medium. Background technique [0002] With the continuous development of computer technology, there are more and more electronic devices capable of capturing images. However, due to various reasons, it often causes the problem that the face area in the captured image or the post-processed image is relatively blurred. There are many reasons for image blurring, such as low hardware configuration of the electronic equipment that captures the image, poor shooting environment, post-processing of the image reduces the resolution of the image, image damage, and so on. [0003] For an image with a small degree of blur in the face region, the related art mainly uses a face deblurring method to process the blurred image. [0004] However, when the blurring of the face area is relatively l...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a model training, image processing method, device, electronic equipment and storage medium to solve the problem that the face deblurring method in the related art has a poor repair effect when repairing an image with severe damage to the face area. The problem

Images