MarkSplatter:基于Splatter Image结构的3D高斯溅射模型通用水印方法
MarkSplatter: Generalizable Watermarking for 3D Gaussian Splatting Model via Splatter Image Structure
AI总结:
针对现有3DGS水印需为每条消息高成本微调的问题,提出首个通用3DGS水印框架MarkSplatter,通过GaussianBridge转换格式、不确定性感知热图保不可感知性、密集分割提取保鲁棒性,单次前向传播即可完成水印嵌入。
摘要:
3D高斯溅射(3D Gaussian Splatting, 3DGS)的日益普及,使得对有效版权保护的需求愈发迫切。当前的3DGS水印方法需要为每条预定义消息执行计算成本高昂的微调流程。我们提出了首个通用水印框架,仅需单次前向传播即可高效保护基于Splatter Image的3DGS模型。我们引入GaussianBridge模块,将非结构化3D高斯转换为Splatter Image格式,从而支持直接通过神经处理嵌入任意消息。为保证水印的不可感知性,我们设计了高斯-不确定性-感知热图预测策略,以保留视觉质量。为实现鲁棒的消息恢复,我们开发了基于密集分割的提取机制,即便水印对象在渲染视图中仅占极小区域,仍能保持可靠的提取效果。项目主页:https://kevinhuangxf.github.io/marksplatter
英文摘要:
The growing popularity of 3D Gaussian Splatting (3DGS) has intensified the need for effective copyright protection. Current 3DGS watermarking methods rely on computationally expensive fine-tuning procedures for each predefined message. We propose the first generalizable watermarking framework that enables efficient protection of Splatter Image-based 3DGS models through a single forward pass. We introduce GaussianBridge that transforms unstructured 3D Gaussians into Splatter Image format, enabling direct neural processing for arbitrary message embedding. To ensure imperceptibility, we design a Gaussian-Uncertainty-Perceptual heatmap prediction strategy for preserving visual quality. For robust message recovery, we develop a dense segmentation-based extraction mechanism that maintains reliable extraction even when watermarked objects occupy minimal regions in rendered views. Project page: https://kevinhuangxf.github.io/marksplatter.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。