X-SG$^2$S: Safe and Generalizable Gaussian Splatting with X-dimensional Watermarks

2025-02-13

AI Summary

The rapid proliferation of 3D Gaussian Splatting (3DGS) digital assets poses urgent challenges for copyright protection. Method: We propose the first lightweight watermarking framework supporting simultaneous embedding and robust extraction of 1D–3D multimodal messages—including text, vectors, and spatial coordinates—by unifying watermark structures across dimensions; designing adaptive gated point sorting and learnable localization mechanisms tightly coupled with 3DGS point features; and introducing XD injection/extraction heads with message chunking and encoding to achieve multimodal watermarking in a single forward pass—without modifying pretrained 3DGS pipelines or parameterization. Contribution/Results: Experiments demonstrate near-lossless rendering quality (PSNR > 35 dB) and strong robustness against geometric perturbations, rendering transformations, and downstream tasks, significantly enhancing the practicality and generality of copyright protection for 3DGS assets.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Security and Privacy: Data transparency and provenance
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

3D Gaussian Splatting (3DGS) has been widely used in 3D reconstruction and 3D generation. Training to get a 3DGS scene often takes a lot of time and resources and even valuable inspiration. The increasing amount of 3DGS digital asset have brought great challenges to the copyright protection. However, it still lacks profound exploration targeted at 3DGS. In this paper, we propose a new framework X-SG$^2$S which can simultaneously watermark 1 to 3D messages while keeping the original 3DGS scene almost unchanged. Generally, we have a X-SG$^2$S injector for adding multi-modal messages simultaneously and an extractor for extract them. Specifically, we first split the watermarks into message patches in a fixed manner and sort the 3DGS points. A self-adaption gate is used to pick out suitable location for watermarking. Then use a XD(multi-dimension)-injection heads to add multi-modal messages into sorted 3DGS points. A learnable gate can recognize the location with extra messages and XD-extraction heads can restore hidden messages from the location recommended by the learnable gate. Extensive experiments demonstrated that the proposed X-SG$^2$S can effectively conceal multi modal messages without changing pretrained 3DGS pipeline or the original form of 3DGS parameters. Meanwhile, with simple and efficient model structure and high practicality, X-SG$^2$S still shows good performance in hiding and extracting multi-modal inner structured or unstructured messages. X-SG$^2$S is the first to unify 1 to 3D watermarking model for 3DGS and the first framework to add multi-modal watermarks simultaneous in one 3DGS which pave the wave for later researches.

Problem

• Addresses copyright protection for 3D Gaussian Splatting assets.
• Proposes a framework for multi-dimensional watermarking in 3DGS.
• Ensures watermarking without altering original 3DGS scene structure.

Innovation

• Multi-modal watermarking in 3DGS
• Self-adaption gate for location selection
• XD-injection heads for message embedding