Research map / 3D vision / 3D generation
Active Neural Mapping
Zike Yan, Haoxiang Yang, Hongbin Zha
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 |
|---|---|---|---|
| identify goal locations by neural variability under weight perturbation instead of FBE frontiers | Active Neural Mapping · Gibson | Completeness (%): 68.91 -> 80.45 | Improved |
| weight-perturbation goal identification vs best baseline on Matterport3D | Active Neural Mapping · Matterport3D | Completeness (%): 71.72 -> 73.15 | Improved |
| replace weight-perturbation uncertainty with MC-Dropout plus BALD (ablation) | Active Neural Mapping · Gibson | MAD (cm): 5.10 -> 5.68 | Hurt |
| replace weight perturbation with MC-Dropout on exploration efficiency (ablation) | Active Neural Mapping · Gibson | Completeness (%): 80.45 -> 79.48 | No clear effect |
| select target candidate by highest variance rather than other criteria (ablation) | Active Neural Mapping · Gibson and MP3D scenes | Completeness (cm) | Improved |
| use softplus MLP with positional embedding instead of ReLU MLP (ablation) | Active Neural Mapping · Gibson | MAD (cm): 6.11 -> 5.10 | Improved |
| use softplus MLP instead of ReLU MLP (ablation) | Active Neural Mapping · Gibson | Completeness (%): 73.70 -> 80.45 | Improved |
| use a different DD-PPO planner checkpoint (Module 3) (ablation) | Active Neural Mapping · Gibson | FPR (%): 28.04 -> 25.09 | Mixed |
| active exploration vs random exploration | Active Neural Mapping · Gibson | MAD (cm): 8.49 -> 5.10 | Improved |
Closest papers by meaning
- NExplore: Exploration with Neural Fields for Autonomous Scene Reconstruction (TPAMI 2026 · similarity 0.88)
- NARUTO: Neural Active Reconstruction from Uncertain Target Observations (CVPR 2024 · similarity 0.87)
- Understanding while Exploring: Semantics-driven Active Mapping (NeurIPS 2025 · similarity 0.86)
- Neural Visibility Field for Uncertainty-Driven Active Mapping (CVPR 2024 · similarity 0.86)
- Swift-Mapping: Online Neural Implicit Dense Mapping in Urban Scenes (AAAI 2024 · similarity 0.85)
- Seeing through Uncertainty: Robust Task-Oriented Optimization in Visual Navigation (NeurIPS 2025 · similarity 0.84)
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.
