Abstract
3D Gaussian Splatting (3DGS) has become a crucial method for acquiring 3D assets. To protect the copyright of these assets, digital watermarking techniques can be applied to embed ownership information discreetly within 3DGS models. However, existing watermarking methods for meshes, point clouds, and implicit radiance fields cannot be directly applied to 3DGS models, as 3DGS models use explicit 3D Gaussians with distinct structures and do not rely on neural networks. Naively embedding the watermark on a pre-trained 3DGS can cause obvious distortion in rendered images. In our work, we propose an uncertainty-based method that constrains the perturbation of model parameters to achieve invisible watermarking for 3DGS. At the message decoding stage, the copyright messages can be reliably extracted from both 3D Gaussians and 2D rendered images even under various forms of 3D and 2D distortions. We conduct extensive experiments on the Blender, LLFF and MipNeRF-360 datasets to validate the effectiveness of our proposed method, demonstrating state-of-the-art performance on both message decoding accuracy and view synthesis quality.
Framework
The overview of our proposed uncertainty-aware 3DGS watermarking. We apply uncertainty estimation to the created 3DGS model. The 3D Gaussians with high uncertainty will be densified. These new densified Gaussians will be regarded as the 3D perturbations and embedded into the original Gaussians to create watermarked 3D Gaussians. The copyright messages can be retrieved from the watermarked 3D Gaussians via the 3d message decoder under various 3D editing. The copyright messages can also be retrieved from the watermarked images via the 2D message decoder against various 2D distortions.