Feature pyramid network for diffusion-based image inpainting detection

Yulan ZHANG, Feng DING, Sam KWONG, Guopu ZHU

Research output: Journal PublicationsJournal Article (refereed)peer-review

26 Citations (Scopus)


Inpainting is a technique that can be employed to tamper with the content of images. In this paper, we propose a novel forensics analysis method for diffusion-based image inpainting based on a feature pyramid network (FPN). Our method features an improved u-shaped net to migrate FPN for multi-scale inpainting feature extraction. In addition, a stagewise weighted cross-entropy loss function is designed to take advantage of both the general loss and the weighted loss to improve the prediction rate of inpainted regions of all sizes. The experimental results demonstrate that the proposed method outperforms several state-of-the-art methods, especially when the size of the inpainted region is small.
Original languageEnglish
Pages (from-to)29-42
JournalInformation Sciences
Early online date23 Apr 2021
Publication statusPublished - Sept 2021
Externally publishedYes

Bibliographical note

This work was supported in part by the National Natural Science Foundation of China under Grant 61872350, Grant 61572489, and Grant 61672443, in part by Hong Kong GRF-RGC General Research Fund under Grant 9042816 (CityU 11209819) and Grant 9042958 (CityU 11203820), in part by the Tip-top Scientific and Technical Innovative Youth Talents of Guangdong Special Support Program under Grant 2019TQ05X696, and in part by the Basic Research Program of Shenzhen under Grant JCYJ20170818163403748.


  • Deep learning
  • Digital forensics
  • Feature pyramid network
  • Image inpainting
  • Tampering detection


Dive into the research topics of 'Feature pyramid network for diffusion-based image inpainting detection'. Together they form a unique fingerprint.

Cite this