Research map / 3D vision / 3D generation
OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perception
Xiaofeng Wang, Zheng Zhu, Wenbo Xu, Yunpeng Zhang, Yi Wei, Xu Chi, Yun Ye, Dalong Du, Jiwen Lu, Xingang 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 adaptive fusion of LiDAR and camera voxel features (ablation) | multi-modal baseline · nuScenes-Occupancy val | mIoU: 14.3 -> 15.1 | Improved |
| combine camera and LiDAR instead of LiDAR only | baseline · nuScenes-Occupancy val | mIoU: 11.7 -> 15.1 | Improved |
| add Cascade Occupancy Network refinement to multi-modal baseline | multi-modal baseline · nuScenes-Occupancy val | mIoU: 15.1 -> 20.1 | Improved |
| use CONet (S=4) instead of high-resolution baseline (S=2) | multi-modal baseline · nuScenes-Occupancy val | mIoU: 19.8 -> 20.1 | Improved |
| sample only 2D semantic features in CONet refinement (ablation) | M-CONet · nuScenes-Occupancy val | mIoU: 15.1 -> 12.1 | Hurt |
| add geometric feature sampling to CONet (ablation) | M-CONet · nuScenes-Occupancy val | mIoU: 15.1 -> 19.3 | Improved |
| add semantic features on top of geometric features (ablation) | M-CONet · nuScenes-Occupancy val | mIoU: 19.3 -> 20.1 | Improved |
| increase camera input size to 1600x900 (ablation) | C-baseline · nuScenes-Occupancy val | mIoU: 8.6 -> 10.3 | Improved |
| replace ResNet50 with ResNet101 (ablation) | C-baseline · nuScenes-Occupancy val | mIoU: 10.3 -> 11.4 | Improved |
| use 10 LiDAR sweeps instead of 1 (ablation) | L-baseline · nuScenes-Occupancy val | mIoU: 11.1 -> 11.7 | Improved |
| reduce stride from 4 to 2 in baseline (ablation) | multi-modal baseline · nuScenes-Occupancy val | mIoU: 15.1 -> 19.8 | Improved |
| adopt surround-view prediction instead of per-view concatenation | camera occupancy · nuScenes-Occupancy val | mIoU: 6.9 -> 10.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MonoScene — MonoScene: Monocular 3D Semantic Scene Completion (CVPR 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
- Collaborative Semantic Occupancy Prediction with Hybrid Feature Fusion in Connected Automated Vehicles (CVPR 2024 · similarity 0.91)
- EvOcc: Accurate Semantic Occupancy for Automated Driving Using Evidence Theory (CVPR 2025 · similarity 0.91)
- UnO: Unsupervised Occupancy Fields for Perception and Forecasting (CVPR 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 0.90)
- OctOcc: High-Resolution 3D Occupancy Prediction with Octree (AAAI 2024 · similarity 0.90)
- SDGOCC: Semantic and Depth-Guided Bird's-Eye View Transformation for 3D Multimodal Occupancy Prediction (CVPR 2025 · similarity 0.90)
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.
