Long Short Term Memory Networks For Light Field View Synthesis

Matthieu Hog 1 Neus Sabater 1 Christine Guillemot 2
2 Sirocco - Analysis representation, compression and communication of visual data
Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : Because light field devices have a limited angular resolution, artificially reconstructing intermediate views is an interesting task. In this work, we propose a novel way to solve this problem using deep learning. In particular, the use of Long Short Term Memory Networks on a plane sweep volume is proposed. The approach has the advantage of having very few parameters and can be run on sequences with arbitrary length. We show that our approach yields results that are competitive with the state-of-the-art for dense light fields. Experimental results also show promising results with light fields with wider baselines.
Document type :
Conference papers
Complete list of metadatas

Cited literature [23 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02202120
Contributor : Christine Guillemot <>
Submitted on : Wednesday, July 31, 2019 - 3:54:16 PM
Last modification on : Friday, August 2, 2019 - 2:24:45 AM

File

LF_view_synthesis_ICIP.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02202120, version 1

Citation

Matthieu Hog, Neus Sabater, Christine Guillemot. Long Short Term Memory Networks For Light Field View Synthesis. IEEE International Conference on Image Processing, Sep 2019, Taipei, Taiwan. ⟨hal-02202120⟩

Share

Metrics

Record views

36

Files downloads

45