Novel Light Field Encoding Framework Based on Optical Flow and Phase Correlation
Iago Storch, Douglas Corrêa, Igor Rosler, Daniel Palomino, Sérgio Bampi, Bruno Zatt · 2022
This work proposes a novel Light Field encoding framework based on optical flow and phase correlation. This encoder works by combining one reference sub-aperture image (SAI) with optical flow and phase correlation information of a few SAIs to synthesize all SAIs of the Light Field. In order to improve the quality of the synthesized Light Field, an HEVC encoder is modified to encode synthesis error and add a quality enhancement layer. The proposed encoder outperforms JPEG Pleno 4D Transform Mode encoder and the HEVC anchor in lower bitrates. Also, it can be easily adapted to the existing multimedia infrastructure.