Light field image editing method and device

Abstract

The invention relates to a light-field image editing method and device. The method comprises the following steps: layering a centre view-angle image in a target scene; respectively generating layer structure masking of a plurality of second sub-images in a centre view angle, and obtaining a plurality of third sub-images in the centre view angle corresponding to the plurality of second sub-images in the centre view angle according to apparent characteristics of the plurality of second sub-images in the centre view angle so as to obtain a second centre view-angle image; determining apparent characteristics of a plurality of second sub-images in other view angles according to the apparent characteristics of the plurality of second sub-images in the centre view angle so as to obtain a plurality of third sub-images in other view angles; and synthesizing the plurality of third sub-images in other view angles so as to obtain a second image in other view angles. According to the invention, structure-keeping editing of a light-field image photographed by a consumer light-field camera can be carried out; and thus, the application range of the consumer light-field camera is further expanded.

Description

technical field [0001] The present invention relates to the technical field of light field image processing, in particular to a light field image editing method and device. Background technique [0002] Light field cameras (also known as plenoptic cameras) have been rapidly developed and popularized in recent years, showing a strong potential to promote technological innovation in the entire field of photography. This kind of camera generally uses a micro-lens array to capture the four-dimensional light field information of the scene, and reconstructs a plenoptic image (Plenoptic Image) containing the apparent features of the same scene taken at different angles. Because it contains intrinsic 3D structural correlation information, light field data provides post-processing effects that traditional 2D images cannot accomplish, such as viewpoint transformation and refocusing. [0003] The hardware of light field cameras is developing very rapidly. At present, Lytro, PiCam and ...

AI Technical Summary

Problems solved by technology

However, light field data can be regarded as a collection of images taken at different angles of the same scene, and the data between different angles of view have a high degree of internal structure correlation, so it is impossible to directly apply the existing two-dimensional image editing methods to light field data. image data

Images