Angular super-resolution reconstruction method on basis of perception loss for images of light fields

Abstract

The invention relates to an angular super-resolution reconstruction method on the basis of perception loss for images of light fields, and belongs to the field of light field imaging. The angular super-resolution reconstruction method includes that mean square errors of image high-dimensional features extracted by pre-training models are used as loss functions, a network model which comprises fourresidual blocks is built, nonlinear mapping relations between observation images and target viewing angle images are learned, and accordingly new viewing angle images can be reconstructed. The angular super-resolution reconstruction method has the advantages that the perception loss for expressing the high-dimensional features is introduced, accordingly, texture details of the super-resolution reconstructed new viewing angle images can be effectively kept, and excellent visual effects can be realized by the angular super-resolution reconstruction method.

Description

technical field [0001] The invention belongs to the field of light field imaging, and relates to a method for super-resolution reconstruction of light field image angles based on perceptual loss. Background technique [0002] Light field imaging has become a focus in the research of next-generation imaging systems. The light field image contains the spatial information and angle information of the light, so the light field camera can capture images from multiple perspectives at one time. Several studies have shown that light field images have good application prospects in many fields, such as image saliency detection, image depth estimation, etc. The basic principle of the full light field camera is to insert a microlens array at the primary image plane of the general imaging system, and the light recorded by each microlens corresponds to the scene image at the same position and different angles of view, so as to obtain a 4-dimensional light field information, including 2 ...

AI Technical Summary

Problems solved by technology

[0006] The above methods estimate the model by minimizing the mean square error between the new perspective image and the target image, but the pixel-level loss is not consistent with the visual loss observed by the human eye. It is possible that the pixel-level loss is higher. more blurred than the one with the lower loss

Images