EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

AI Watermark Robustness: Evaluating Methods to Identify AI-Generated Content After Editing or Compression

Rohit Kumar, Swarna Verma, Deepanshu Mittal, Armaan Singh

International Journal of Research and Review in Applied Science, Humanities, and Technology · 26.12.2024 · 142-147

DOI: 10.71143/qrzzwv89 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Research and Review in Applied Science, Humanities, and Technology, online erschienen am 26.12.2024. Verfasst von Rohit Kumar und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The explosion of Generative Artificial Intelligence (AI) systems that produce photorealistic images, natural-sounding audio, coherent video and fluent text has intensified the demand for dependable mechanisms to trace data back to its source to establish a provenance of synthetic content. One of the most used technical solutions to this requirement is digital watermarking, which consists of adding a slight and imperceptible signal to media produced. In the real world, however, very little content can be presented to a verifier without some alterations: it is often cropped, re-sized, re-encoded, filtered, paraphrased, or deliberately attacked to remove identifying signals. In this paper, we provide a systematic survey of image, video, text, and audio AI watermarking methods, focusing on their resilience to post-generation editing and lossy compression. We categorise approaches into two types of embedding: post-hoc and in-generation and review the threat models and evaluation metrics employed to assess robustness and then compare representative techniques such as hybrids in the frequency domain, learned encoder-decoder-based approaches, diffusion process watermarks, and token-level language model watermarks. We also explore attacks, such as adversarial removal, regeneration and forgery attacks, which are specifically targeting the watermark persistence, and finally collate benchmark results that show a persistent trade-off between robustness and quality/capacity. It ends with open problems including the lack of standard robustness measures, vulnerability to surrogate generative attacks, and the absence of cross-modal or multilingual robustness guarantees, as well as some further directions that are promising, such as semantically based watermarking and certified robustness frameworks.

Zitieren

APA

Kumar, R., Verma, S., Mittal, D., & Singh, A. (2024). AI Watermark Robustness: Evaluating Methods to Identify AI-Generated Content After Editing or Compression. International Journal of Research and Review in Applied Science, Humanities, and Technology, 142-147. https://doi.org/10.71143/qrzzwv89

Vancouver

Kumar R, Verma S, Mittal D, Singh A. AI Watermark Robustness: Evaluating Methods to Identify AI-Generated Content After Editing or Compression. International Journal of Research and Review in Applied Science, Humanities, and Technology. 2024:142-147. doi:10.71143/qrzzwv89

BibTeX
@article{Kumar2024,
  author = {Kumar, Rohit and Verma, Swarna and Mittal, Deepanshu and Singh, Armaan},
  title = {AI Watermark Robustness: Evaluating Methods to Identify AI-Generated Content After Editing or Compression},
  journal = {International Journal of Research and Review in Applied Science, Humanities, and Technology},
  year = {2024},
  pages = {142--147},
  doi = {10.71143/qrzzwv89}
}
Kostenlos starten →