By the same authors

From the same journal

From the same journal

Single Image Super Resolution via Neighbor Reconstruction

Research output: Contribution to journalArticle

Author(s)

  • Zhihong Zhang
  • Chen Xu
  • Zhonghao Zhang
  • Guo Chen
  • Yide Cai
  • Zeli Wang
  • Heng Li
  • Edwin R. Hancock

Department/unit(s)

Publication details

JournalPattern Recognition Letters
DateAccepted/In press - 22 Apr 2019
DateE-pub ahead of print (current) - 22 Apr 2019
Early online date22/04/19
Original languageUndefined/Unknown

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.

Bibliographical note

This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.

    Research areas

  • Manifold learning, Neighbor Reconstruction, Super Resolution

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations