Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Occupancy Anticipation for Efficient Exploration and Navigation
Santhosh K. Ramakrishnan, Ziad Al-Halah, Kristen Grauman
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 occupancy anticipation (OccAnt) to depth mapper | ANS(depth) · Gibson val per-frame | mean IoU %: 19.3 -> 56.1 | Improved |
| add occupancy anticipation (OccAnt) to rgb mapper | ANS(rgb) · Gibson val per-frame | mean IoU %: 13.5 -> 46.1 | Improved |
| add RGB to depth input (rgbd instead of depth) | OccAnt · Gibson val per-frame | mean IoU %: 56.1 -> 56.5 | No clear effect |
| add anticipation reward (ablation) | OccAnt(rgbd) · Gibson small, noisy exploration | map accuracy (m^2): 16.9 -> 22.7 | Improved |
| add anticipation reward (ablation) | OccAnt(depth) · Gibson large, noisy exploration | map accuracy (m^2): 44.2 -> 50.3 | Improved |
| add anticipation reward (ablation) | OccAnt(rgbd) · Matterport3D, noisy exploration | map accuracy (m^2): 76.3 -> 99.9 | Improved |
| add anticipation reward (ablation) | OccAnt(rgb) · Gibson large, noisy exploration | map accuracy (m^2): 44.2 -> 42.1 | Hurt |
| add anticipation reward (ablation) | OccAnt(depth) · Matterport3D, noise-free exploration | map accuracy (m^2): 104.7 -> 96.5 | Hurt |
| add anticipation reward (ablation) | OccAnt(depth) · Gibson val PointNav | SPL %: 78.6 -> 77.8 | No clear effect |
| add anticipation reward (ablation) | OccAnt(rgbd) · Gibson val PointNav | SPL %: 77.9 -> 80.0 | Improved |
| add occupancy anticipation to rgb mapper | ANS(rgb) · Gibson val PointNav | SPL %: 66.8 -> 71.2 | Improved |
| add occupancy anticipation to depth mapper | ANS(depth) · Matterport3D, noisy exploration | map accuracy (m^2): 72.5 -> 94.1 | Improved |
Closest papers by meaning
- Spatiotemporal Decoupling for Efficient Vision-Based Occupancy Forecasting (CVPR 2025 · similarity 0.88)
- COME: Adding Scene-Centric Forecasting Control to Occupancy World Model (NeurIPS 2025 · similarity 0.86)
- AdaOcc: Adaptive 3D Occupancy Prediction for Embodied Tasks (NeurIPS 2026 · similarity 0.86)
- Cam4DOcc: Benchmark for Camera-Only 4D Occupancy Forecasting in Autonomous Driving Applications (CVPR 2024 · similarity 0.86)
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.86)
- Versatile Navigation Under Partial Observability via Value-guided Diffusion Policy (CVPR 2024 · similarity 0.86)
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.
