Plenoptic image compression via simplified subaperture projection

Abstract

In this paper, a generalized encoding architecture is proposed for compressing plenoptic images to reduce the redundancy among the subaperture images. Based on the homography analysis, simplification of subaperture projection is performed due to high correlations among adjacent views. Then, a pseudo-video sequence consisting of central subaperture images and a sequence consisting of residual images between the central image and the adjacent images are generated by the proposed partitioning and reordering methods. They are compressed by temporal coding tools in HEVC. The experimental results demonstrate that the proposed method outperforms state-of-the-art pseudo-video compression methods by an average of 25.3%/27.4%/21.7%/36.7%/5.3% bitrate saving with comparable/lower computational complexity.