Focal Stack Based Light Field Coding for Refocusing Applications

Since a light field camera generates huge volume of image data, it requires powerful compression for efficient transmission and storage of its data. In its refocusing applications, users want to see their objects of interest at different depth levels. To address this, we first render a focal stack that contains multi-focus images, then compress it instead of original light field data. By doing so, we can minimize the amount of data to process for its target refocusing applications. Our experiment results show that the proposed method can save 54.09% of bitrate and have PSNR gain of 2.08 dB at the same bitrate.