Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Frontier-enhanced Topological Memory with Improved Exploration Awareness for Embodied Visual Navigation
Xinru Cui, Qiming Liu, Zhe Liu, Hesheng Wang
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 ghost nodes (Fg) to topological memory (ablation) | FTM · Gibson image-goal nav (overall) | SR: 0.537 -> 0.625 | Improved |
| predict ghost node features with online NeRF (Fn) (ablation) | FTM · Gibson image-goal nav (hard) | SR: 0.376 -> 0.451 | Improved |
| add Multi-Stage Memory Extraction (MSME) (ablation) | FTM · Gibson image-goal nav (overall) | SR: 0.549 -> 0.625 | Improved |
| use NeRF-rendered ghost features instead of previous-view features (ablation) | FTM · Gibson image-goal nav (overall) | SR: 0.583 -> 0.625 | Improved |
| use NeRF-rendered ghost features instead of SE3DS GAN (ablation) | FTM · Gibson image-goal nav (overall) | SR: 0.613 -> 0.625 | Improved |
| replace NeRF ghost features with ground-truth observations (ablation) | FTM · Gibson image-goal nav (overall) | SR: 0.625 -> 0.649 | Improved |
| use FTM instead of FGPrompt in easy episodes | image-goal navigation · Gibson easy | SR: 0.796 -> 0.774 | Hurt |
| use FTM instead of best baseline overall | image-goal navigation · Gibson overall | SR: 0.587 -> 0.625 | Improved |
Closest papers by meaning
- MemoNav: Working Memory Model for Visual Navigation (CVPR 2024 · similarity 0.90)
- Semi-parametric topological memory for navigation (ICLR 2018 · similarity 0.89)
- 3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning (CVPR 2025 · similarity 0.86)
- Change-Robust Online Topological Memory for Long-Term Relocalization and Semantic Navigation (NeurIPS 2026 · similarity 0.86)
- Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator (NeurIPS 2026 · similarity 0.86)
- Planning from Imagination: Episodic Simulation and Episodic Memory for Vision-and-Language Navigation (AAAI 2025 · 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.
