EUVIMEDEUVIMED entdecken
Crossref

T2IW: Joint Text to Image & Watermark Generation

Guokai Zhang, Yuting Su, Lanjun Wang, Guochang Liu, Dan Song, An-An Liu

Machine Intelligence Research · 2026-08 · 23 (4) · 804-822

DOI: 10.1007/s11633-025-1606-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machine Intelligence Research, online erschienen am 11.06.2026. Verfasst von Guokai Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 76 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent developments in text-conditioned image generative models have revolutionized the production of realistic results. Unfortunately, this has also led to an increase in privacy violations and the spread of false information, which requires the need for traceability, privacy protection, and other security measures. However, existing text-to-image paradigms lack the technical capabilities to link traceable messages with image generation. In this study, we introduce a novel task for the joint generation of text to image and watermark (T2IW), and develop a framework to generalize to text-conditioned image generative models. This T2IW framework is employed to ensure minimal damage to image quality when generating a compound image by forcing the semantic feature and the watermark signal to be compatible in pixels. Additionally, by utilizing principles from Shannon information theory and non-cooperative game theory, we are able to separate the revealed image and the revealed watermark from the compound image. Furthermore, we strengthen the watermark robustness of our approach by subjecting the compound image to various post-processing attacks, with minimal pixel distortion observed in the revealed watermark. Extensive experiments have demonstrated the remarkable achievements of our framework in image quality, watermark invisibility, and watermark robustness, supported by our proposed set of evaluation metrics.

Zitieren

APA

Zhang, G., Su, Y., Wang, L., Liu, G., Song, D., & Liu, A. (2026). T2IW: Joint Text to Image & Watermark Generation. Machine Intelligence Research, 23(4), 804-822. https://doi.org/10.1007/s11633-025-1606-9

Vancouver

Zhang G, Su Y, Wang L, Liu G, Song D, Liu A. T2IW: Joint Text to Image & Watermark Generation. Machine Intelligence Research. 2026;23(4):804-822. doi:10.1007/s11633-025-1606-9

BibTeX
@article{Zhang2026,
  author = {Zhang, Guokai and Su, Yuting and Wang, Lanjun and Liu, Guochang and Song, Dan and Liu, An-An},
  title = {T2IW: Joint Text to Image & Watermark Generation},
  journal = {Machine Intelligence Research},
  year = {2026},
  volume = {23},
  number = {4},
  pages = {804--822},
  doi = {10.1007/s11633-025-1606-9}
}
Kostenlos starten →