# Light Field Compact Representation
_by Xiaoran Jiang (IETR VAADER) - 2022.12.01_

###### tags: `VAADER` `Seminar`

## Video
{%youtube oYa0ZkjYexw %}

## Abstract

Light fields represent very large volumes of high-dimensional data. Finding effective but compact representations of light fields has therefore become a key challenge for practical use of this technology. Some methods exploiting light field prior to obtain compact representations will be introduced in this presentation. In particular, a homography-based low rank approximation method and a deep learning method using an untrained generative model. These methods achieve competitive performance in light field compression and inverse problems. A learning-based view synthesis method for light field will be also presented.