A Dunhuang mural restoration network based on mask guidance and Transformer architecture.

In the process of Dunhuang mural restoration, models often struggle to focus on damaged areas, leading to issues such as color deviation and blurred lines. To address these problems, we propose a Dunhuang mural restoration network based on mask guidance and Transformer architecture, named MGTNet. Fi...

Full description

Bibliographic Details
Published in:Digital Scholarship in the Humanities Vol. 40; no. 3; pp. 1051 - 1064
Main Authors: Zhang, Qiang, Ren, Doudou, Gao, Ying, Zhang, Yixuan, Chen, Tao
Format: Article
Published: Oxford University Press / USA Sep2025
Subjects:
Online Access:View this record in EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=188027895&site=ehost-live
header:
  @attributes:
    shortDbName: hlh
    uiTerm: 188027895
    longDbName: Humanities International Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        2055768X
        JEO9
      jtl: Digital Scholarship in the Humanities
      issn: 2055768X
      maglogo: N
    pubinfo:
      dt: Sep2025
      vid: 40
      iid: 3
      pid: 622
      pub: Oxford University Press / USA
    artinfo:
      ui:
        188027895
        10.1093/llc/fqaf044
      ppf: 1051
      ppct: 13
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
              size: 2.4MB
      tig:
        atl: A Dunhuang mural restoration network based on mask guidance and Transformer architecture.
      aug:
        au:
          Zhang, Qiang
          Ren, Doudou
          Gao, Ying
          Zhang, Yixuan
          Chen, Tao
        affil:
          School of Liberal Arts, Huaiyin Normal University, Huaian 223300, China
          School of Humanities and Social Development, Nanjing Agricultural University, Nanjing 210095, China
          School of Computer Science and Technology, Xinjiang University, Urumqi 830017, China
          School of Cultural Heritage and Information Management, Shanghai University, Shanghai 200444, China
          School of Information Management, Sun Yat-sen University, Guangzhou 510006, China
      su:
        Art conservation & restoration
        Transformer models
        Artificial intelligence
        Digital humanities
        Image reconstruction
        Cultural property
        Image enhancement (Imaging systems)
        Dunhuang (China)
      sug:
        subj:
          Dunhuang (China)
          Art conservation & restoration
          Transformer models
          Artificial intelligence
          Digital humanities
          Image reconstruction
          Cultural property
          Image enhancement (Imaging systems)
      keyword:
        adaptive attention
        deep learning
        digital humanities
        Dunhuang mural restoration
        mask guidance
        Transformer
      ab: In the process of Dunhuang mural restoration, models often struggle to focus on damaged areas, leading to issues such as color deviation and blurred lines. To address these problems, we propose a Dunhuang mural restoration network based on mask guidance and Transformer architecture, named MGTNet. First, we design a mural focus attention module to dynamically optimize both channel and spatial information in an adaptive manner. Next, we introduce a mural enhancement module that leverages the long-range dependency capturing capability of Transformers to improve restoration quality. Finally, a mask-guided downsampling module is proposed, which fuses the mask image as prior knowledge with downsampled features, enhancing the model's ability to perceive damaged areas. Experiments on a publicly available Dunhuang mural dataset demonstrate that the proposed method outperforms comparison algorithms in terms of objective evaluation metrics such as PSNR and SSIM, validating the effectiveness of the algorithm. This study demonstrates the practical application potential of the algorithm in Dunhuang mural restoration and contributes to the significant advancements of artificial intelligence in the field of digital humanities.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: Y
      custom: © 2019 EADH: The European Association for Digital Humanities.
      item: Digital Scholarship in the Humanities
      holder: Oxford University Press / USA
      dt:
        @attributes:
          year: 2025
    holdings:
      @attributes:
        islocal: N