paper-with-me

홈 › Papers

StealthAttack: Robust 3D Gaussian Splatting Poisoning via Density-Guided Illusions

2025-10-02 · Bo-Hsu Ke, You-Zhe Xie, Yu-Lun Liu, Wei-Chen Chiu arxiv

3D scene representation methods like Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have significantly advanced novel view synthesis. As these methods become prevalent, addressing their vulnerabilities becomes critical. We analyze 3DGS robustness against image-level poisoning attacks and propose a novel density-guided poisoning method. Our method strategically injects Gaussian points into low-density regions identified via Kernel Density Estimation (KDE), embedding viewpoint-dependent illusory objects clearly visible from poisoned views while minimally affecting innocent views. Additionally, we introduce an adaptive noise strategy to disrupt multi-view consistency, further enhancing attack effectiveness. We propose a KDE-based evaluation protocol to assess attack difficulty systematically, enabling objective benchmarking for future research. Extensive experiments demonstrate our method's superior performance compared to state-of-the-art techniques. Project page: https://hentci.github.io/stealthattack/

📄 PDF Abstract BibTeX arXiv:2510.02314

Code (0)

등록된 구현이 없습니다.

Tasks

Novel View SynthesisDensity Estimation

Similar Papers 제목 키워드 기반

Shape-guided Gaussian Splatting for Sparse-View X-ray 3D Reconstruction

2026-09-09 · Pranav Poudel, Florence Dell'Aniello Picard, Nairouz Shehata, Frédéric Lavoie 외 arxiv

Sparse-view X-ray 3D reconstruction is essential for reducing radiation exposure, but recovering a density field from a handful of X-ray projections is severely ill-posed. Recently, 3D Gaussian Splatting has achieved sta…

3D Reconstruction

GDGS: 3D Gaussian Splatting Via Geometry-Guided Initialization And Dynamic Density Control

2025-07-01 · Xingjun Wang, Lianlei Shan arxiv

We propose a method to enhance 3D Gaussian Splatting (3DGS)~\cite{Kerbl2023}, addressing challenges in initialization, optimization, and density control. Gaussian Splatting is an alternative for rendering realistic image…

D$^2$GS: Depth-and-Density Guided Gaussian Splatting for Stable and Accurate Sparse-View Reconstruction

2025-10-09 · Meixi Song, Xin Lin, Dizhe Zhang, Haodong Li 외 arxiv

Recent advances in 3D Gaussian Splatting (3DGS) enable real-time, high-fidelity novel view synthesis (NVS) with explicit 3D representations. However, performance degradation and instability remain significant under spars…

Novel View Synthesis

2DGS-Room: Seed-Guided 2D Gaussian Splatting with Geometric Constrains for High-Fidelity Indoor Scene Reconstruction

2024-12-04 · Wanting Zhang, Haodong Xiang, Zhichao Liao, Xiansong Lai 외

The reconstruction of indoor scenes remains challenging due to the inherent complexity of spatial structures and the prevalence of textureless regions. Recent advancements in 3D Gaussian Splatting have improved novel vie…

Indoor Scene ReconstructionNovel View SynthesisSurface Reconstruction

GaussianBody: Clothed Human Reconstruction via 3d Gaussian Splatting

2024-01-18 · Mengtian Li, Shengxiang Yao, Zhifeng Xie, Keyu Chen

In this work, we propose a novel clothed human reconstruction method called GaussianBody, based on 3D Gaussian Splatting. Compared with the costly neural radiance based models, 3D Gaussian Splatting has recently demonstr…