A Neural Network with Adversarial Loss for Light Field Synthesis from a Single Image

Simon Evain (1) , Christine Guillemot (1)

1 Sirocco - Analysis representation, compression and communication of visual data

This paper describes a lightweight neural network architecture with an adversarial loss for generating a full light field from one single image. The method is able to estimate disparity maps and automatically identify occluded regions from one single image thanks to a disparity confidence map based on forward-backward consistency checks. The disparity confidence map also controls the use of an adversarial loss for occlusion handling. The approach outperforms reference methods when trained and tested on light field data. Besides, we also designed the method so that it can efficiently generate a full light field from one single image, even when trained only on stereo data. This allows us to generalize our approach for view synthesis to more diverse data and semantics.

Mots clés: Monocular; View synthesis; Deep learning; Light field; Depth estimation

Domaines: Traitement du signal et de l'image [eess.SP]

Simon Evain, Christine Guillemot. A Neural Network with Adversarial Loss for Light Field Synthesis from a Single Image. VISAPP 2021 - 16th International Conference on Computer Vision Theory and Applications, Feb 2021, Vienna, Austria. pp.1-10. ⟨hal-03024210v1⟩