3D Scene Representation

Follow this topic Ask about this topic

3D scene representation studies how to encode the shape, appearance, meaning, and motion of environments so they can be reconstructed, rendered, edited, or used by robots and other AI systems. Common approaches include explicit representations such as meshes, voxels, point clouds, Gaussian splats, and scene graphs, alongside implicit neural fields and compact learned tokens; these are used in tasks from novel-view synthesis to dynamic mapping and scene generation. Current work increasingly combines geometric structure with language or semantic features, and tests representations on reconstruction benchmarks as well as navigation, spatial reasoning, and manipulation tasks. A central tension is balancing visual fidelity and geometric reliability against memory, computation, and scalability, especially for large, changing scenes and incomplete observations.

Latest papers

242 papers35 papers in the last four weeks, up 35% on the four weeks before. 0.2% of all new papers.

GaussTrap: Stealthy Backdoor Attacks on 3D Gaussian Splatting for Targeted Scene Misperception
Apr 29, 2025Jiaxin Hong, Sixu Chen, Shuoyang Sun +53D Gaussian Splatting 3DGS Training
Harbin Institute of Technology(Shenzhen) · South China University of Technology · Shenzhen Internation Graduate School, Tsinghua University +1
NumGrad-Pull: Numerical Gradient Guided Tri-plane Representation for Surface Reconstruction from Point Clouds
Nov 26, 2024Ruikai Cui, Binzhu Xie, Shi Qiu +3Implicit Neural Representations 3D Reconstruction
School of Computing, The Australian National University, Australia · Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong SAR, China · Department of Computer Science & Software Engineering, the University of Western Australia, Australia

Apr 29, 2025J/K move · Enter open · S save

GaussTrap: Stealthy Backdoor Attacks on 3D Gaussian Splatting for Targeted Scene Misperception

Jiaxin Hong, Sixu Chen, Shuoyang Sun, Hongyao Yu+4

Harbin Institute of Technology(Shenzhen) · South China University of Technology · Shenzhen Internation Graduate School, Tsinghua University+1

3D Gaussian Splatting 3DGS Training 3D Scene Representation

Open paper AskPDF arXiv

3D Gaussian Splatting (3DGS) has recently emerged as a powerful paradigm for real-time scene representation and novel view synthesis, gaining traction in safety-critical applications such as autonomous driving and AR/VR systems. However, the security vulnerabilities of 3DGS remain largely unexplored. In this paper, we conduct a comprehensive study of backdoor threats in 3DGS pipelines. We uncover an inherent security vulnerability in the adaptive density control mechanism of 3DGS, originally designed for detailed reconstruction, which enables attackers to induce localized gaussian proliferation, effectively embedding malicious backdoors into the scene geometry while remaining invisible from other viewpoints. To expose this threat, we propose GaussTrap, a novel stealthy poisoning framework tailored for 3DGS. GaussTrap operates in three stages (adversarial injection, geometric stabilization, and global fidelity restoration) to embed imperceptible yet adversarial Gaussian components into the reconstruction. It ensures consistent poisoning effects at specific viewpoints while preserving photorealism in benign views, thereby achieving high attack efficacy with low detectability. Extensive experiments on both synthetic and real-world benchmarks demonstrate that GaussTrap reliably induces targeted visual corruption under trigger conditions while maintaining high rendering fidelity elsewhere, highlighting a critical security blind spot in modern 3D rendering systems. The code is available at https://github.com/acang425/GaussTrap.

Latest papers

GaussTrap: Stealthy Backdoor Attacks on 3D Gaussian Splatting for Targeted Scene Misperception

NumGrad-Pull: Numerical Gradient Guided Tri-plane Representation for Surface Reconstruction from Point Clouds