PROJECT TITLE :

Spatial and Angular Resolution Enhancement of Light Fields Using Convolutional Neural Networks - 2018

ABSTRACT:

Light field imaging extends the ancient photography by capturing both spatial and angular distribution of sunshine, which permits new capabilities, together with post-capture refocusing, post-capture aperture control, and depth estimation from one shot. Micro-lens array (MLA) based mostly light field cameras supply a cost-effective approach to capture light field. A major disadvantage of MLA based mostly lightweight field cameras is low spatial resolution, which is because of the fact that a single image sensor is shared to capture each spatial and angular info. During this Project, we have a tendency to present a learning based mostly light-weight field enhancement approach. Both spatial and angular resolution of captured light-weight field is enhanced using convolutional neural networks. The proposed method is tested with real light field knowledge captured with a Lytro light field camera, clearly demonstrating spatial and angular resolution improvement.