Zhang, Zhihong, Xu, Chen, Zhang, Zhonghao et al. (5 more authors) (2019) Single Image Super Resolution via Neighbor Reconstruction. Pattern Recognition Letters. ISSN 0167-8655
Abstract
Super Resolution (SR) is a complex, ill-posed problem where the aim is to construct the mapping between the low and high resolution manifolds of image patches. Anchored neighborhood regression for SR (namely A+ [27]) has shown promising results. In this paper we present a new regression-based SR algorithm that overcomes the limitations of A+ and benefits from an innovative and simple Neighbor Reconstruction Method (NRM). This is achieved by vector operations on an anchored point and its corresponding neighborhood. NRM reconstructs new patches which are closer to the anchor point in the manifold space. Our method is robust to NRM sparsely-sampled points: increasing PSNR by 0.5 dB compared to the next best method. We comprehensively validate our technique on standardised datasets and compare favourably with the state-of-the-art methods: we obtain PSNR improvement of up to 0.21 dB compared to previously-reported work.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. |
Keywords: | Manifold learning, Neighbor Reconstruction, Super Resolution |
Dates: |
|
Institution: | The University of York |
Academic Units: | The University of York > Faculty of Sciences (York) > Computer Science (York) |
Depositing User: | Pure (York) |
Date Deposited: | 24 Apr 2019 10:10 |
Last Modified: | 16 Oct 2024 15:38 |
Published Version: | https://doi.org/10.1016/j.patrec.2019.04.021 |
Status: | Published online |
Refereed: | Yes |
Identification Number: | 10.1016/j.patrec.2019.04.021 |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:145309 |
Download
Filename: 1_s2.0_S0167865519301369_main.pdf
Description: Single Image Super Resolution via Neighbor Reconstruction
Licence: CC-BY-NC-ND 2.5