Research map / 3D vision / 3D generation
Occupancy Learning with Spatiotemporal Memory
Ziyang Leng, Jiawei Yang, Wenlong Yi, Bolei Zhou
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 spatiotemporal memory and memory attention (vs no temporal fusion) (ablation) | FB-OCC with ST-Occ · Occ3D | mIoU: 37.39 -> 41.17 | Improved |
| add dynamic awareness to memory attention (ablation) | ST-Occ · Occ3D | mIoU: 41.17 -> 41.73 | Improved |
| add uncertainty awareness to memory attention (ablation) | ST-Occ · Occ3D | mIoU: 41.17 -> 41.85 | Improved |
| add both uncertainty and dynamic awareness (ablation) | ST-Occ · Occ3D | mIoU: 41.17 -> 42.13 | Improved |
| add historical class activation to uncertainty estimation (ablation) | ST-Occ · Occ3D | mIoU: 41.17 -> 41.45 | Improved |
| add feature similarity to uncertainty estimation (ablation) | ST-Occ · Occ3D | mIoU: 41.45 -> 41.73 | Improved |
| add averaged log variance to uncertainty estimation (ablation) | ST-Occ · Occ3D | mIoU: 41.73 -> 41.85 | Improved |
| add occupancy flow to full uncertainty estimation (ablation) | ST-Occ · Occ3D | mIoU: 41.85 -> 42.13 | Improved |
| add uncertainty awareness to memory attention (ablation) | ST-Occ · Occ3D | mSTCV: 9.25 -> 8.85 | Improved |
| replace FB-OCC temporal fusion with ST-Occ | FB-OCC · Occ3D | mSTCV: 12.18 -> 8.68 | Improved |
| use unified temporal modeling instead of stacked (ablation) | ST-Occ · Occ3D | Fusion time (ms): 84 -> 24 | Improved |
| use unified temporal modeling instead of recurrent (ablation) | ST-Occ · Occ3D | Fusion time (ms): 705 -> 24 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- OccFormer — OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Prediction (ICCV 2023) · better on 1 of 1 shared result
- ViewFormer — ViewFormer: NeRF-Free Neural Rendering from Few Images Using Transformers (ECCV 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Spatiotemporal Decoupling for Efficient Vision-Based Occupancy Forecasting (CVPR 2025 · similarity 0.92)
- ODG: Occupancy Prediction Using Dual Gaussians (NeurIPS 2025 · similarity 0.91)
- STCOcc: Sparse Spatial-Temporal Cascade Renovation for 3D Occupancy and Scene Flow Prediction (CVPR 2025 · similarity 0.90)
- Achieving Speed-Accuracy Balance in Vision-based 3D Occupancy Prediction via Geometric-Semantic Disentanglement (AAAI 2025 · similarity 0.90)
- OctOcc: High-Resolution 3D Occupancy Prediction with Octree (AAAI 2024 · similarity 0.90)
- CSV-Occ: Fusing Multi-frame Alignment for Occupancy Prediction with Temporal Cross State Space Model and Central Voting Mechanism (ICML 2025 · similarity 0.89)
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.
