Incremental image restoration method based on wireframe and edge structure

Abstract

The invention relates to an incremental image restoration method based on a wireframe and an edge structure. The method comprises the following steps: acquiring a scene data picture; constructing a masking layer adapted to a downstream task to carry out model training; constructing a structure recovery model and training the structure recovery model; constructing an up-sampling network of the wireframe structure and training the up-sampling network; if the resolution of the masked image is greater than 256 * 256, performing up-sampling on a 256 * 256 repair wireframe and an edge structure by using a structure up-sampling network until the resolution of the repair wireframe and the edge structure is the same as that of the masked image; inputting the repaired wireframe and edge information into a structural feature encoder to obtain structural features; obtaining a covering position code according to the covering layer; constructing an image inpainting network and training the image inpainting network; and after the model training is finished, image restoration is carried out. Compared with the prior art, the method has the advantages of good image restoration effect, high adaptability and the like.

Description

technical field [0001] The invention relates to the technical field of image restoration, in particular to an incremental image restoration method based on wireframe and edge structures. Background technique [0002] Image inpainting is to solve the problem of filling missing areas in damaged images, and the purpose is to ensure the authenticity and rationality of the texture details of the repaired image areas. It has been studied for many years as a long-term method. Image inpainting is extremely useful for many real-world applications such as photography, object removal, image editing, etc. [0003] As a research hotspot in computer vision in recent years, there have been many valuable works on image inpainting algorithms based on deep learning. Many of these works are devoted to improving the texture details of filled regions through the improvement of the model. For example, the document "Resolution-robust large mask inpainting with fourier convolutions" proposes an i...

AI Technical Summary

Problems solved by technology

However, the edge information in the above-mentioned document "Edgeconnect: Generative image inpainting with adversarial edge learning" is often based on gradient calculations, so there are some shortcomings, such as straight-line edges are often intermittent, and the structure of some color-insensitive regions is lost due to dependence on gradients information

The structure in the document "Learning a sketch tensor space for image inpainting of man-madescenes" is based on adversarial training to recover, and the recovery effect is poor, and it trains the entire network from scratch to integrate structural information, which is very expensive and cannot cope with high Resolved Image Repair Issues

Images