Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis

Abstract: 

The recent years have witnessed the widespread of light field imaging in interactive and immersive visual applications. To record the directional information of the light rays, larger storage space is required by light field images compared with conventional 2D images. Hence, the efficient compression of light field image is highly desired for further applications. In this paper, we propose a novel light field image compression scheme using multi- branch spatial transformer networks based view synthesis. Firstly, a sparse subset of views are selected and are rearranged into a pseudo sequence to be encoded by an video codec at encoder. Then the other unselected views are synthesized based on the similarity between neighboring views with our proposed method at decoder. To better characterize the non-linear relationship between the sub-views, a multi-branch spatial transformer networks (MSTN) is designed to adaptively learn the affine transformations between the neighboring views, which are used to warp the input views to generate accurate approximation of the target views. Moreover, to better obtain the final view by the generated approximation views, the Wasserstein generative adversarial networks(WGAN) is applied with the improved training. Experimental results show the superior compression performance of our scheme compared with the state-of-the-art methods.

up
0 users have voted:

Paper Details

Authors:
Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin
Submitted On:
28 March 2020 - 3:09am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin
Session:
Posters
Document Year:
2020
Cite

Document Files

DCC 2020 poster.pdf

(64)

Subscribe

[1] Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin, "Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5041. Accessed: Jul. 11, 2020.
@article{5041-20,
url = {http://sigport.org/5041},
author = {Jin Wang; Qianwen Wang; Ruiqin Xiong; Qing Zhu; Baocai Yin },
publisher = {IEEE SigPort},
title = {Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis},
year = {2020} }
TY - EJOUR
T1 - Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis
AU - Jin Wang; Qianwen Wang; Ruiqin Xiong; Qing Zhu; Baocai Yin
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5041
ER -
Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin. (2020). Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis. IEEE SigPort. http://sigport.org/5041
Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin, 2020. Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis. Available at: http://sigport.org/5041.
Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin. (2020). "Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis." Web.
1. Jin Wang, Qianwen Wang, Ruiqin Xiong, Qing Zhu, Baocai Yin. Light Field Image Compression Using Multi-Branch Spatial Transformer Networks Based View Synthesis [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5041