Research map / 3D vision / Depth and stereo
Learning To Explore Using Active Neural SLAM
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the modular, hierarchical Active Neural SLAM approach instead of the strongest end-to-end baseline | Exploration · Gibson Val | % Cov.: 0.789 -> 0.948 | Helped |
| Use the modular, hierarchical Active Neural SLAM approach instead of the strongest end-to-end baseline | Exploration · MP3D Test | % Cov.: 0.378 -> 0.521 | Helped |
| Transfer ANS trained on Exploration to PointGoal without additional training | ANS · Gibson Val | Success: 0.827 -> 0.950 | Helped |
| Add a learned Local Policy instead of using a deterministic planner-following policy (ablation) | Active Neural SLAM · Gibson Val Overall | % Cov.: 0.941 -> 0.948 | Helped |
| Add a learned Global Policy instead of frontier-based exploration (ablation) | Active Neural SLAM · Gibson Val Large | % Cov.: 0.782 -> 0.862 | Helped |
| Add a learned Pose Estimator instead of using sensor readings as the pose estimate (ablation) | Active Neural SLAM · Gibson Val Large | % Cov.: 0.771 -> 0.862 | Helped |
| Add a learned Pose Estimator instead of using sensor readings as the pose estimate (ablation) | Active Neural SLAM · Gibson Val Small | % Cov.: 0.973 -> 0.983 | Helped |
Closest papers by meaning
- Learning Exploration Policies for Navigation (ICLR 2019 · similarity 0.88)
- Versatile Navigation Under Partial Observability via Value-guided Diffusion Policy (CVPR 2024 · similarity 0.86)
- Learning to Navigate Efficiently and Precisely in Real Environments (CVPR 2024 · similarity 0.86)
- NextBestPath: Efficient 3D Mapping of Unseen Environments (ICLR 2025 · similarity 0.86)
- MNE-SLAM: Multi-Agent Neural SLAM for Mobile Robots (CVPR 2025 · similarity 0.86)
- Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction (ICLR 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.