GS-Marker: Generalizable and Robust Watermarking for 3D Gaussian Splatting

2025-03-24

AI Summary

To address copyright protection needs for 3D models in the generative AI era, this paper introduces the first single-forward-pass, universal, and robust watermarking framework for 3D Gaussian Splatting (3DGS). Existing methods suffer from scene-specific iterative optimization, poor generalizability, and vulnerability to distortions induced by free-viewpoint rendering. To overcome these limitations, we propose a differentiable architecture comprising a 3D encoder, a distortion layer, and a 2D decoder—integrating differentiable rendering, adversarial distortion modeling, and adaptive parameter perturbation to circumvent gradient blocking in the renderer. Our method achieves cross-scene generalization after a single training phase. Experimental results demonstrate significantly improved watermark extraction accuracy, superior model fidelity compared to iterative approaches, an order-of-magnitude reduction in computational latency, and robust watermark retrieval under arbitrary viewing angles.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web search models and ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

In the Generative AI era, safeguarding 3D models has become increasingly urgent. While invisible watermarking is well-established for 2D images with encoder-decoder frameworks, generalizable and robust solutions for 3D remain elusive. The main difficulty arises from the renderer between the 3D encoder and 2D decoder, which disrupts direct gradient flow and complicates training. Existing 3D methods typically rely on per-scene iterative optimization, resulting in time inefficiency and limited generalization. In this work, we propose a single-pass watermarking approach for 3D Gaussian Splatting (3DGS), a well-known yet underexplored representation for watermarking. We identify two major challenges: (1) ensuring effective training generalized across diverse 3D models, and (2) reliably extracting watermarks from free-view renderings, even under distortions. Our framework, named GS-Marker, incorporates a 3D encoder to embed messages, distortion layers to enhance resilience against various distortions, and a 2D decoder to extract watermarks from renderings. A key innovation is the Adaptive Marker Control mechanism that adaptively perturbs the initially optimized 3DGS, escaping local minima and improving both training stability and convergence. Extensive experiments show that GS-Marker outperforms per-scene training approaches in terms of decoding accuracy and model fidelity, while also significantly reducing computation time.

Problem

• Develop robust watermarking for 3D Gaussian Splatting models
• Ensure generalization across diverse 3D models and distortions
• Enable efficient single-pass training without per-scene optimization

Innovation

• Single-pass watermarking for 3D Gaussian Splatting
• Adaptive Marker Control for stable training
• Distortion layers enhance watermark resilience