Search papers, labs, and topics across Lattice.
This paper addresses the significant gap in video protection against both image-to-video and fine-tuning-based customization by introducing Temporally Consistent Universal Adversarial Perturbations (TC-UAP). The method overcomes three critical challenges in video protection: the vulnerability of frame-by-frame perturbations to temporal compression, the ineffectiveness of single video perturbations against unseen videos, and the inconsistency of perturbations under temporal attacks. Empirical results demonstrate that TC-UAP provides superior identity protection compared to existing methods while maintaining robustness against various temporal attacks.
TC-UAP is the first method to effectively safeguard videos from both reference- and tuning-based customization, achieving unprecedented identity protection.
Recent diffusion-based video generation models have enabled high-quality personalized video customization through both tuning-based pipelines, which fine-tune a video diffusion model, and reference-based pipelines such as image-to-video generation. However, these capabilities raise serious concerns about personal privacy, identity ownership and intellectual property protection. Existing anti-customization works focus on protecting images, while protection for videos against both reference- and tuning-based customization remains largely underexplored. Protecting videos in this setting raises three challenges: (i) Image-level perturbations, optimized frame by frame, cannot survive temporal compression by 3D video VAE. (ii) A video-level perturbation optimized on a single video is vulnerable to temporal editing and fails to protect unseen videos. (iii) Temporally inconsistent perturbations are not robust to temporal attacks. To address these challenges, we propose Temporally Consistent Universal Adversarial Perturbations (TC-UAP), the first protection method against both reference- and tuning-based video customization. TC-UAP optimizes an identity-level multi-frame UAP over sliding windows from multiple videos, accounting for local temporal dependencies induced by temporal compression in video VAE and enabling a single perturbation to protect unseen videos of varying lengths. Moreover, we introduce intrinsic temporal modeling and an extrinsic surrogate temporal-attack loss, which make the perturbation temporally consistent and robust to unseen temporal attacks. Empirically, quantitative and qualitative results show that TC-UAP achieves the strongest identity protection compared with existing methods under both reference- and tuning-based video customization, and remains robust under multiple unseen temporal attacks.
