RDSplat: Robust Watermarking Against Diffusion Editing for 3D Gaussian Splatting

2025-12-07

AI Summary

To address the vulnerability of 3D Gaussian Splatting (3DGS) watermarks to erasure under diffusion-based editing, this paper proposes RDSplat—the first watermarking framework for 3DGS explicitly designed for robustness against diffusion editing. Our method exploits the inherent preservation of low-frequency Gaussian components during diffusion editing, embedding watermarks into the low-frequency subspaces of covariance matrices and opacities. To achieve end-to-end robust optimization directly on native 3DGS representations, we introduce multi-domain collaborative regularization—including covariance constraints, 2D frequency-domain filtering, and a low-pass Gaussian blur surrogate model—combined with adversarial fine-tuning. Extensive experiments across three benchmark datasets demonstrate that RDSplat maintains watermark imperceptibility while achieving significantly superior robustness against diverse diffusion-editing attacks compared to state-of-the-art methods.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Adversarial Learning & Robustness
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Security and Privacy: Data transparency and provenance
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

3D Gaussian Splatting (3DGS) has enabled the creation of digital assets and downstream applications, underscoring the need for robust copyright protection via digital watermarking. However, existing 3DGS watermarking methods remain highly vulnerable to diffusion-based editing, which can easily erase embedded provenance. This challenge highlights the urgent need for 3DGS watermarking techniques that are intrinsically resilient to diffusion-based editing. In this paper, we introduce RDSplat, a Robust watermarking paradigm against Diffusion editing for 3D Gaussian Splatting. RDSplat embeds watermarks into 3DGS components that diffusion-based editing inherently preserve, achieved through (i) proactively targeting low-frequency Gaussians and (ii) adversarial training with a diffusion proxy. Specifically, we introduce a multi-domain framework that operates natively in 3DGS space and embeds watermarks into diffusion-editing-preserved low-frequency Gaussians via coordinated covariance regularization and 2D filtering. In addition, we exploit the low-pass filtering behavior of diffusion-based editing by using Gaussian blur as an efficient training surrogate, enabling adversarial fine-tuning that further enhances watermark robustness against diffusion-based editing. Empirically, comprehensive quantitative and qualitative evaluations on three benchmark datasets demonstrate that RDSplat not only maintains superior robustness under diffusion-based editing, but also preserves watermark invisibility, achieving state-of-the-art performance.

Problem

• Develops robust watermarking for 3D Gaussian Splatting against diffusion editing
• Embeds watermarks in low-frequency Gaussians preserved during editing
• Uses adversarial training to enhance watermark resilience to editing

Innovation

• Embeds watermarks into low-frequency Gaussians for resilience
• Uses adversarial training with diffusion proxy for robustness
• Applies coordinated covariance regularization and 2D filtering