Safe-VAR: Safe Visual Autoregressive Model for Text-to-Image Generative Watermarking

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Ziyi, Tan, Songbai, Xu, Gang, Qiu, Xuerui, Xu, Hongbin, Meng, Xin, Li, Ming, Yu, Fei Richard
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910875858763776
author Wang, Ziyi
Tan, Songbai
Xu, Gang
Qiu, Xuerui
Xu, Hongbin
Meng, Xin
Li, Ming
Yu, Fei Richard
author_facet Wang, Ziyi
Tan, Songbai
Xu, Gang
Qiu, Xuerui
Xu, Hongbin
Meng, Xin
Li, Ming
Yu, Fei Richard
contents With the success of autoregressive learning in large language models, it has become a dominant approach for text-to-image generation, offering high efficiency and visual quality. However, invisible watermarking for visual autoregressive (VAR) models remains underexplored, despite its importance in misuse prevention. Existing watermarking methods, designed for diffusion models, often struggle to adapt to the sequential nature of VAR models. To bridge this gap, we propose Safe-VAR, the first watermarking framework specifically designed for autoregressive text-to-image generation. Our study reveals that the timing of watermark injection significantly impacts generation quality, and watermarks of different complexities exhibit varying optimal injection times. Motivated by this observation, we propose an Adaptive Scale Interaction Module, which dynamically determines the optimal watermark embedding strategy based on the watermark information and the visual characteristics of the generated image. This ensures watermark robustness while minimizing its impact on image quality. Furthermore, we introduce a Cross-Scale Fusion mechanism, which integrates mixture of both heads and experts to effectively fuse multi-resolution features and handle complex interactions between image content and watermark patterns. Experimental results demonstrate that Safe-VAR achieves state-of-the-art performance, significantly surpassing existing counterparts regarding image quality, watermarking fidelity, and robustness against perturbations. Moreover, our method exhibits strong generalization to an out-of-domain watermark dataset QR Codes.
format Preprint
id arxiv_https___arxiv_org_abs_2503_11324
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Safe-VAR: Safe Visual Autoregressive Model for Text-to-Image Generative Watermarking
Wang, Ziyi
Tan, Songbai
Xu, Gang
Qiu, Xuerui
Xu, Hongbin
Meng, Xin
Li, Ming
Yu, Fei Richard
Multimedia
Computer Vision and Pattern Recognition
Image and Video Processing
With the success of autoregressive learning in large language models, it has become a dominant approach for text-to-image generation, offering high efficiency and visual quality. However, invisible watermarking for visual autoregressive (VAR) models remains underexplored, despite its importance in misuse prevention. Existing watermarking methods, designed for diffusion models, often struggle to adapt to the sequential nature of VAR models. To bridge this gap, we propose Safe-VAR, the first watermarking framework specifically designed for autoregressive text-to-image generation. Our study reveals that the timing of watermark injection significantly impacts generation quality, and watermarks of different complexities exhibit varying optimal injection times. Motivated by this observation, we propose an Adaptive Scale Interaction Module, which dynamically determines the optimal watermark embedding strategy based on the watermark information and the visual characteristics of the generated image. This ensures watermark robustness while minimizing its impact on image quality. Furthermore, we introduce a Cross-Scale Fusion mechanism, which integrates mixture of both heads and experts to effectively fuse multi-resolution features and handle complex interactions between image content and watermark patterns. Experimental results demonstrate that Safe-VAR achieves state-of-the-art performance, significantly surpassing existing counterparts regarding image quality, watermarking fidelity, and robustness against perturbations. Moreover, our method exhibits strong generalization to an out-of-domain watermark dataset QR Codes.
title Safe-VAR: Safe Visual Autoregressive Model for Text-to-Image Generative Watermarking
topic Multimedia
Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2503.11324