Research map / 3D vision / Gaussian splatting
GaussNav: Gaussian Splatting for Visual Navigation
Xiaohan Lei, Min Wang, Wengang Zhou, Houqiang Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add classifier (ablation) | GaussNav · HM3D IIN | Success: 0.375 -> 0.725 | Improved |
| add Match module (ablation) | GaussNav · HM3D IIN | Success: 0.444 -> 0.725 | Improved |
| add novel view synthesis (ablation) | GaussNav · HM3D IIN | SPL: 0.557 -> 0.578 | Improved |
| use DISK+LightGlue instead of SIFT+FLANN (ablation) | GaussNav · HM3D IIN | Success: 0.655 -> 0.725 | Improved |
| use DISK+LightGlue instead of GlueStick (ablation) | GaussNav · HM3D IIN | Success: 0.723 -> 0.725 | No clear effect |
| add classifier to local matching (ablation) | GaussNav matching · Matching success | time and success: 0.039 higher success without classifier | Mixed |
| increase rendered views from n_v=3 to 5 vertically (ablation) | GaussNav NVS · Matching success | Success: 0.832 -> 0.831 | No clear effect |
| add NVS to matching (original to horizontal n_v=3) (ablation) | GaussNav · Matching success | Success: 0.811 -> 0.824 | Improved |
| use ground truth match (ablation) | GaussNav · HM3D IIN | Success: 0.725 -> 0.850 | Improved |
| use ground truth goal localization on top of GT match (ablation) | GaussNav · HM3D IIN | Success: 0.850 -> 0.946 | Improved |
| replace episodic map with scene-specific map | IEVE InternImage · HM3D IIN | SPL: 0.252 -> 0.347 | Improved |
| use Semantic Gaussian map instead of scene-specific BEV map | GaussNav vs IEVE InternImage · HM3D IIN | SPL: 0.347 -> 0.578 | Improved |
Closest papers by meaning
- VTGaussian-SLAM: RGBD SLAM for Large Scale Scenes with Splatting View-Tied 3D Gaussians (ICML 2025 · similarity 0.88)
- Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator (NeurIPS 2026 · similarity 0.88)
- HUGS: Holistic Urban 3D Scene Understanding via Gaussian Splatting (CVPR 2024 · similarity 0.87)
- Memorize What Matters: Emergent Scene Decomposition from Multitraverse (NeurIPS 2024 · similarity 0.87)
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · similarity 0.87)
- MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
