Can Protective Watermarking Safeguard the Copyright of 3D Gaussian Splatting?

2025-11-27

AI Summary

This work exposes critical security vulnerabilities in existing 3D Gaussian Splatting (3DGS) watermarking schemes for copyright protection, providing the first systematic validation of their susceptibility to purification attacks. To address this, we propose GSPure—the first watermark purification method tailored to 3DGS representations. GSPure models watermark distribution via view-dependent rendering analysis and employs geometry-aware Gaussian clustering coupled with primitive contribution decoupling to precisely localize and losslessly remove watermarks. Experiments demonstrate that GSPure reduces watermark PSNR by 16.34 dB while degrading the original scene’s PSNR by less than 1 dB. It significantly outperforms prior methods in removal strength, reconstruction fidelity, and cross-scene generalizability. Our work establishes a new paradigm for security evaluation and defense of 3DGS watermarking, bridging a fundamental gap between watermark robustness and practical deployability in neural 3D representations.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Web data integration and cleaning

Abstract

3D Gaussian Splatting (3DGS) has emerged as a powerful representation for 3D scenes, widely adopted due to its exceptional efficiency and high-fidelity visual quality. Given the significant value of 3DGS assets, recent works have introduced specialized watermarking schemes to ensure copyright protection and ownership verification. However, can existing 3D Gaussian watermarking approaches genuinely guarantee robust protection of the 3D assets? In this paper, for the first time, we systematically explore and validate possible vulnerabilities of 3DGS watermarking frameworks. We demonstrate that conventional watermark removal techniques designed for 2D images do not effectively generalize to the 3DGS scenario due to the specialized rendering pipeline and unique attributes of each gaussian primitives. Motivated by this insight, we propose GSPure, the first watermark purification framework specifically for 3DGS watermarking representations. By analyzing view-dependent rendering contributions and exploiting geometrically accurate feature clustering, GSPure precisely isolates and effectively removes watermark-related Gaussian primitives while preserving scene integrity. Extensive experiments demonstrate that our GSPure achieves the best watermark purification performance, reducing watermark PSNR by up to 16.34dB while minimizing degradation to original scene fidelity with less than 1dB PSNR loss. Moreover, it consistently outperforms existing methods in both effectiveness and generalization.

Problem

• Evaluates vulnerabilities in 3D Gaussian Splatting watermarking for copyright protection.
• Proposes a specialized purification framework to remove watermarks from 3DGS assets.
• Ensures robust watermark removal while preserving original scene fidelity.

Innovation

• Introduces GSPure, first purification framework for 3DGS watermarking
• Uses view-dependent rendering analysis and geometric feature clustering
• Isolates and removes watermark primitives while preserving scene integrity