Research map / 3D vision / 3D generation
Fast-BEV: A Fast and Strong Bird’s-Eye View Perception Baseline
Yangguang Li, Bin Huang, Zeren Chen, Yufeng Cui, Feng Liang, Mingzhu Shen, Fenggang Liu, Enze Xie, Lu Sheng, Wanli Ouyang, Jing Shao
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 image augmentation (ablation) | Fast-BEV 1F · nuScenes val | mAP: 0.247 -> 0.285 | Improved |
| add BEV augmentation (ablation) | Fast-BEV 1F · nuScenes val | mAP: 0.247 -> 0.265 | Improved |
| add both image and BEV augmentation (ablation) | Fast-BEV 1F · nuScenes val | mAP: 0.247 -> 0.293 | Improved |
| add one history frame (2F) (ablation) | Fast-BEV · nuScenes val | NDS: 0.373 -> 0.451 | Improved |
| add history frames up to 4F (ablation) | Fast-BEV · nuScenes val | NDS: 0.451 -> 0.466 | Improved |
| add image multi-scale only (ablation) | Fast-BEV 2F · nuScenes val | mAP: 0.321 -> 0.320 | No clear effect |
| add BEV multi-scale only (ablation) | Fast-BEV 2F · nuScenes val | mAP: 0.321 -> 0.316 | Hurt |
| add image and BEV multi-scale together (ablation) | Fast-BEV 2F · nuScenes val | NDS: 0.451 -> 0.455 | Improved |
| increase image resolution from 256x448 to 832x1440 (ablation) | Fast-BEV · nuScenes val | NDS: 0.419 -> 0.491 | Improved |
| increase voxel resolution beyond 200x200x6 (ablation) | Fast-BEV · nuScenes val | mAP: 0.352 -> 0.337 | No clear effect |
| scale 2D encoder from R18 to R101 (ablation) | Fast-BEV 4F · nuScenes val | mAP: 0.293 -> 0.345 | Improved |
| scale 3D encoder from 2 to 6 blocks (ablation) | Fast-BEV 4F · nuScenes val | NDS: 0.446 -> 0.451 | 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.
- BEVDepth — BEVDepth: Acquisition of Reliable Depth for Multi-View 3D Object Detection (AAAI 2023) · better on 2 of 2 shared results
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
- Delving Into the Devils of Bird’s-Eye-View Perception: A Review, Evaluation and Recipe (TPAMI 2023 · similarity 0.93)
- M-BEV: Masked BEV Perception for Robust Autonomous Driving (AAAI 2024 · similarity 0.92)
- HotBEV: Hardware-oriented Transformer-based Multi-View 3D Detector for BEV Perception (NeurIPS 2023 · similarity 0.92)
- Improving Bird's Eye View Semantic Segmentation by Task Decomposition (CVPR 2024 · similarity 0.90)
- Toward Real-world BEV Perception: Depth Uncertainty Estimation via Gaussian Splatting (CVPR 2025 · similarity 0.89)
- A Black-Box Evaluation Framework for Semantic Robustness in Bird’s Eye View Detection (AAAI 2025 · similarity 0.88)
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.
