为视频生成模型设计高容量鲁棒水印，可精准定位篡改帧
VIDSTAMP: A Temporally-Aware Watermark for Ownership and Integrity in Video Diffusion Models
- 通过双阶段微调解码器，在潜空间嵌入每帧48比特消息
- 水印视觉影响小，抗常见编辑操作，检测准确率达0.96
- 支持动态水印和帧级篡改定位，适合版权保护场景
视频扩散模型能生成逼真且时间一致的视频，引发溯源、所有权与完整性问题。水印可通过直接嵌入内容来缓解这些问题。理想水印需具备足够容量、不可感知性及对常见视频操作的鲁棒性。现有方法在容量、推理开销或画质上存在不足。本文提出VidStamp，一种基于潜在空间的视频水印框架，通过微调视频扩散模型的解码器，在每帧中嵌入帧级信息。第一阶段使用静态图像数据集促进空间消息分离；第二阶段利用合成视频序列恢复时间一致性。该方法实现高容量水印（每帧48比特），同时保持低感知影响。VidStamp还支持通过控制信号在推理时动态选择消息模板，增加通信通道灵活性。我们在Stable Video Diffusion (I2V)、OpenSora和Wan (T2V) 上评估，结果表明其相比VideoSeal、VideoShield和RivaGAN具有更低的log P值和更强可检测性，帧级水印设计使篡改定位准确率达0.96，优于VideoShield基线。
原文摘要 · Abstract (English)
Video diffusion models can generate realistic and temporally consistent videos. This raises concerns about provenance, ownership, and integrity. Watermarking can help address these issues by embedding metadata directly into the content. To work well, a watermark needs enough capacity for meaningful metadata. It must also stay imperceptible and remain robust to common video manipulations. Existing methods struggle with limited capacity, extra inference cost, or reduced visual quality. We introduce VidStamp, a watermarking framework that embeds frame-level messages through the decoder of a latent video diffusion model. The decoder is fine-tuned in two stages. The first stage uses static image datasets to encourage spatial message separation. The second stage uses synthesized video sequences to restore temporal consistency. This approach enables high-capacity watermarks with minimal perceptual impact. VidStamp also supports dynamic watermarking through a control signal that selects message templates during inference. This adds flexibility and creates a second channel for communication. We evaluate VidStamp on Stable Video Diffusion (I2V), OpenSora, and Wan (T2V). The system embeds 48 bits per frame while preserving visual quality and staying robust to common distortions. Compared with VideoSeal, VideoShield, and RivaGAN, it achieves lower log P-values and stronger detectability. Its frame-wise watermarking design also enables precise temporal tamper localization, with an accuracy of 0.96, which exceeds the VideoShield baseline. Code: https://github.com/SPIN-UMass/VidStamp
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。