Research map / 3D vision / Gaussian splatting
IGL-Nav: Incremental 3D Gaussian Localization for Image-goal Navigation
Wenxuan Guo, Xiuwei Xu, Hang Yin, Ziwei Wang, Jianjiang Feng, Jie Zhou, Jiwen Lu
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 |
|---|---|---|---|
| use higher subdivision level 2 instead of 1 for rotation discretization (ablation) | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 19.7 -> 41.3 | Improved |
| use subdivision level 3 instead of 2 (ablation) | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 41.3 -> 57.0 | Improved |
| use subdivision level 3 instead of 2 (ablation) | IGL-Nav · Free-view image-goal, Wide FOV | SPL (%): 42.1 -> 55.0 | Improved |
| add rendering-based stopper with fine localization (ablation) | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 45.7 -> 57.0 | Improved |
| add rendering-based stopper with fine localization (ablation) | IGL-Nav · Free-view image-goal, Wide FOV | SR (%): 46.2 -> 63.3 | Improved |
| use 3DGS stopper and matching-constrained optimization instead of SLING (ablation) | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 49.0 -> 57.0 | Improved |
| use 3DGS stopper and matching-constrained optimization instead of SLING (ablation) | IGL-Nav · Free-view image-goal, Wide FOV | SPL (%): 45.0 -> 55.0 | Improved |
| add SLING stopper instead of none (ablation) | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 45.7 -> 49.0 | Improved |
| use measured depth instead of predicted depth (ablation) | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 53.8 -> 57.0 | Improved |
| use measured depth instead of predicted depth (ablation) | IGL-Nav · Free-view image-goal, Wide FOV | SR (%): 61.0 -> 63.3 | Improved |
| train on free-view image-goal data instead of zero-shot transfer | IGL-Nav · Free-view image-goal, Narrow FOV | SR (%): 43.1 -> 57.0 | Improved |
| use IGL-Nav instead of RNR-Map | IGL-Nav · Gibson image-goal, Straight overall | SPL (%): 43.9 -> 64.1 | Improved |
Closest papers by meaning
- VIGS-SLAM: Visual Inertial Gaussian Splatting SLAM (ECCV 2026 · similarity 0.86)
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.86)
- VTGaussian-SLAM: RGBD SLAM for Large Scale Scenes with Splatting View-Tied 3D Gaussians (ICML 2025 · similarity 0.85)
- MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM (CVPR 2025 · similarity 0.85)
- Gaussian Splatting SLAM (CVPR 2024 · similarity 0.85)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.85)
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.
