Light field editing in the gradient domain

Abstract

Abstract This paper presents a new method for light field applications such as content replacement and fusion in the gradient domain. This approach is inspired by successful gradient domain based image and video editing techniques. A necessary and important part of gradient‐based solutions is recovering the signal of interest from artificially generated, and typically non‐integrable, gradient data. As such, a new algorithm is developed to reconstruct a light field from a given gradient data set. In the algorithm, first, the 4D Haar wavelet decomposition of the light field is obtained from the given gradient data. Then, the light field is obtained from a wavelet synthesis step. This algorithm is intended as a building block for gradient‐based light field editing methods, and as such, its performance is analysed on a set of benchmark light field data sets. The proposed reconstruction algorithm is an essential part in developing solutions for two light field problems: light field editing and light field fusion. Results show that processing light fields in the gradient domain offers significant advantages over processing in the intensity domain.

Keywords

• Optical, image and video signal processing
• Computer vision and image processing techniques
• Video signal processing
• Graphics techniques