PRIME: Protect Your Videos From Malicious Editing

Guanlin Li, Shuai Yang, Jie Zhang, Han Qiu, Tianwei Zhang

Keywords: Diffusion Model, Video Protection, Video Editing

Abstract: Over the years, video generation has experienced significant advancement. A variety of open-source models emerge, making it surprisingly easy to manipulate and edit videos with just a few simple prompts. While these cutting-edge technologies have gained huge popularity, they have also given rise to concerns regarding the privacy and portrait rights of individuals: malicious users can exploit these tools for deceptive or illegal purposes. Existing works on protecting images against generative models cannot be directly grafted to video protection, due to their efficiency and effectiveness limitations. Motivated by this, we introduce PRIME, a new methodology dedicated to the protection of videos from unauthorized editing via generative models. Our key idea is to craft highly transferable and robust perturbations, which can be efficiently added to the protected videos to disrupt their editing feasibility. We perform comprehensive evaluations using both objective metrics and human studies. The results indicate that PRIME only needs 8.3% GPU hours of existing state-of-the-art methods while achieving better protection results.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 1017