Research map / Detection and segmentation / 3D detection and driving
PolarFormer: Multi-Camera 3D Object Detection with Polar Transformer
Yanqin Jiang, Li Zhang, Zhenwei Miao, Xiatian Zhu, Jin Gao, Weiming Hu, Yu-Gang Jiang
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 |
|---|---|---|---|
| Use Polar features and Polar predictions rather than Cartesian features and predictions (ablation) | PolarFormer · nuScenes val | mAP: 38.1 -> 39.6 | Improved |
| Use Polar features while retaining Cartesian predictions (ablation) | PolarFormer · nuScenes val | mAP: 38.1 -> 38.5 | Improved |
| Decode predictions in Polar rather than Cartesian coordinates (ablation) | PolarFormer with Polar features · nuScenes val | mAP: 38.5 -> 39.6 | Improved |
| Use fixed Sine PE rather than 2D learnable PE (ablation) | PolarFormer · nuScenes val | mAP: 38.8 -> 39.6 | Improved |
| Use fixed Sine PE rather than 3D PE (ablation) | PolarFormer · nuScenes val | mAP: 38.5 -> 39.6 | Improved |
| Add multi-scale Polar BEV representation instead of single-scale (ablation) | PolarFormer · nuScenes val | mAP: 39.1 -> 39.6 | Improved |
| Add multi-scale BEV representation (ablation) | Cartesian-feature PolarFormer-CC · nuScenes val | mAP: 38.1 -> 38.1 | No clear effect |
| Set azimuth resolution to 256 rather than 240 (ablation) | PolarFormer · nuScenes val | mAP: 38.8 -> 39.6 | Improved |
| Set azimuth resolution to 256 rather than 272 (ablation) | PolarFormer · nuScenes val | mAP: 38.8 -> 39.6 | Improved |
| Set radius resolution to 64 rather than 56 (ablation) | PolarFormer · nuScenes val | mAP: 38.7 -> 39.6 | Improved |
| Set radius resolution to 64 rather than 72 (ablation) | PolarFormer · nuScenes val | mAP: 38.5 -> 39.6 | Improved |
| Set radius resolution to 64 rather than 80 (ablation) | PolarFormer · nuScenes val | mAP: 38.7 -> 39.6 | Improved |
Later papers that compared with this method
4 later accepted papers report PolarFormer as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- MemDistill: Distilling LiDAR Knowledge into Memory for Camera-Only 3D Object Detection (ICCV 2025 · better on 2 of 2 shared results)
- FrustumFormer: Adaptive Instance-Aware Resampling for Multi-View 3D Detection (CVPR 2023 · better on 4 of 4 shared results)
- X3KD: Knowledge Distillation Across Modalities, Tasks and Stages for Multi-Camera 3D Object Detection (CVPR 2023 · better on 3 of 4 shared results)
- Object as Query: Lifting Any 2D Object Detector to 3D Detection (ICCV 2023 · better on 4 of 4 shared results)
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.
- BEVFormer — BEVFormer: Learning Bird’s-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers (ECCV 2022) · better on 5 of 7 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
- PARTNER: Level Up the Polar Representation for 3D Object Detection (IJCV 2026 · similarity 0.90)
- PARTNER: Level up the Polar Representation for LiDAR 3D Object Detection (ICCV 2023 · similarity 0.88)
- BEVFormer: Learning Bird’s-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers (ECCV 2022 · similarity 0.87)
- A Versatile Multi-View Framework for LiDAR-Based 3D Object Detection With Guidance From Panoptic Segmentation (CVPR 2022 · similarity 0.87)
- BEVFormer: Learning Bird’s-Eye-View Representation From LiDAR-Camera via Spatiotemporal Transformers (TPAMI 2024 · similarity 0.87)
- Towards Generalizable Multi-Camera 3D Object Detection via Perspective Rendering (AAAI 2025 · 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.
