Light field coding based on flexible view ordering for unfocused plenoptic camera images
Cristian Perra · UNICA IRIS Institutional Research Information System (University of Cagliari) · 2017
Plenoptic cameras are devices designed for sampling the light field entering their main lens. The raw information captured by the sensor can then be processed for rendering views at different focal distance or from different perspective. A plenoptic camera generates images with high resolution and high redundant information if compared with a conventional camera having resolution equal to that of a light field rendered view. A light field compression architecture designed for coding views obtained from unfocused plenoptic images is proposed in this paper. The intent of this proposal is to provide a lossy compression tool enabling access to group of light field views. The experimental analysis aims at exploring the performance of the proposed method. The obtained results show that the proposed flexible ordering approach can provide advanced control to the user device in selecting the set of views that are effectively required for a given rendering application.