CSCGAN: Cross-Space Contrastive Learning for Blind Image Inpainting
Applied Sciences · 19.03.2026 · 16 (6) · 2969
DOI: 10.3390/app16062969 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 19.03.2026. Verfasst von Sheng Jin und 7 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Zhejiang Province Program, Zhejiang Provincial Cultural Relics Protection Science and Technology Project, National Program of China.
Abstract (Originalsprache)
Existing general image inpainting works require the user to customize a mask to indicate the region to be inpainted. However, the mask is often hard to calibrate accurately in real-world applications, e.g., graffiti removal. Blind image inpainting aims to automatically restore the degraded image into the visually reasonable one without a priori mask to indicate the area to be repaired. So far, most proposed blind inpainting methods convert the task into general inpainting by predicting the mask before inpainting. However, these methods are highly dependent on mask prediction results, which may produce inferior inpainting results if the prediction is inaccurate. To address this issue, we propose a two-stage blind inpainting framework with two novel designs: (1) cross-space contrastive learning, to remove the noise in the degraded images and realize the automatic inpainting in the latent space by reducing the distance of the degraded images and the corresponding complete images in the latent space; and (2) mask-aware adversarial training, to minimize the mutual information between the inpainted feature and the noise. Extensive experiments prove that our blind inpainting framework performs better on multiple datasets than the state-of-the-art methods.
Förderung
- Zhejiang Province Program
- Zhejiang Provincial Cultural Relics Protection Science and Technology Project
- National Program of China
- Macau project: Key technology research and display system development for new personalized controllable dressing dynamic display
- Ningbo Science and Technology Plan Project
Zitieren
APA
Jin, S., Zhang, W., Chu, T., Zhang, Z., Zhao, L., Xing, W., Lin, H., & Chen, L. (2026). CSCGAN: Cross-Space Contrastive Learning for Blind Image Inpainting. Applied Sciences, 16(6), 2969. https://doi.org/10.3390/app16062969
Vancouver
Jin S, Zhang W, Chu T, Zhang Z, Zhao L, Xing W, et al. CSCGAN: Cross-Space Contrastive Learning for Blind Image Inpainting. Applied Sciences. 2026;16(6):2969. doi:10.3390/app16062969
BibTeX
@article{Jin2026,
author = {Jin, Sheng and Zhang, Weijing and Chu, Tianyi and Zhang, Zhanjie and Zhao, Lei and Xing, Wei and Lin, Huaizhong and Chen, Lixia},
title = {CSCGAN: Cross-Space Contrastive Learning for Blind Image Inpainting},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {6},
pages = {2969},
doi = {10.3390/app16062969}
}
EUVIMED