Home

Awesome

LFattNet: Attention-based View Selection Networks for Light-field Disparity Estimation

<img src="/images/paper_concept.jpg" width="555" align=center />

Attention-based View Selection Networks for Light-field Disparity Estimation

Yu-Ju Tsai,<sup>1</sup> Yu-Lun Liu,<sup>1,2</sup> Ming Ouhyoung,<sup>1</sup> Yung-Yu Chuang<sup>1</sup>
<sup>1</sup>National Taiwan University, <sup>2</sup>MediaTek

AAAI Conference on Artificial Intelligence (AAAI), Feb 2020

Network Architecture

Network Architecture

SOTA on 4D Light Field Benchmark

<img src="/images/benchmark.PNG" width="555" align=center />

Environment

Ubuntu            16.04
Python            3.5.2
Tensorflow-gpu    1.10
CUDA              9.0.176
Cudnn             7.1.4

Train LFattNet

Evaluate LFattNet

Citation

@inproceedings{Tsai:2020:ABV,
        author = {Tsai, Yu-Ju and Liu, Yu-Lun and Ouhyoung, Ming and Chuang, Yung-Yu},
        title = {Attention-based View Selection Networks for Light-field Disparity Estimation},
        booktitle = {Proceedings of the 34th Conference on Artificial Intelligence (AAAI)},
        year = {2020}
}

Last modified data: 2020/09/14.

The code is modified and heavily borrowed from EPINET: https://github.com/chshin10/epinet