Deep Recurrent Network for Fast and Full-Resolution Light Field Deblurring
- Authors
- Lumentut, Jonathan Samuel; Kim, Tae Hyun; Ramamoorthi, Ravi; Park, In Kyu
- Issue Date
- Dec-2019
- Publisher
- IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- Keywords
- Recurrent network; light field image; blind deblurring; dataset; 6-DOF motion
- Citation
- IEEE SIGNAL PROCESSING LETTERS, v.26, no.12, pp.1788 - 1792
- Indexed
- SCIE
SCOPUS
- Journal Title
- IEEE SIGNAL PROCESSING LETTERS
- Volume
- 26
- Number
- 12
- Start Page
- 1788
- End Page
- 1792
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/146673
- DOI
- 10.1109/LSP.2019.2947379
- ISSN
- 1070-9908
- Abstract
- The popularity of parallax-based image processing is increasing while in contrast early works on recovering sharp light field from its blurry input (deblurring) remain stagnant. State-of-the-art blind light field deblurring methods suffer from several problems such as slow processing, reduced spatial size, and simplified motion blur model. In this paper, we solve these challenging problems by proposing a novel light field recurrent deblurring network that is trained under 6 degree-of-freedom camera motion-blur model. By combining the real light field captured using Lytro Illum and synthetic light field rendering of 3D scenes from UnrealCV, we provide a large-scale blurry light field dataset to train the network. The proposed method outperforms the state-of-the-art methods in terms of deblurring quality, the capability of handling full-resolution, and a fast runtime.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 컴퓨터소프트웨어학부 > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/146673)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.