Research map / Detection and segmentation / Open-set and few-shot detection
DA-BEV: Unsupervised Domain Adaptation for Bird's Eye View Perception
Kai Jiang, Jiaxing Huang, Weiying Xie, Jie Lei, Yunsong Li, Ling Shao, Shijian Lu
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 query-based adversarial learning on image-view features (ablation) | PETR · Daytime->Nighttime | mAP: 15.93 -> 16.45 | Improved |
| add BEV QAL alongside image-view QAL (ablation) | DA-BEV · Daytime->Nighttime | mAP: 16.45 -> 16.96 | Improved |
| add co-regularization between image-view and BEV QAL (ablation) | DA-BEV · Daytime->Nighttime | mAP: 16.96 -> 17.80 | Improved |
| add image-view query-based self-training (ablation) | DA-BEV · Daytime->Nighttime | mAP: 17.80 -> 18.53 | Improved |
| add BEV query-based self-training (ablation) | DA-BEV · Daytime->Nighttime | mAP: 18.53 -> 19.39 | Improved |
| add co-regularization of QST between image-view and BEV (ablation) | DA-BEV · Daytime->Nighttime | mAP: 19.39 -> 20.27 | Improved |
| pseudo-label ratio 10% instead of 20% (ablation) | QST · Daytime->Nighttime | mAP: 20.27 -> 18.24 | Hurt |
| vary pseudo-label ratio between 15% and 30% (ablation) | QST · Daytime->Nighttime | mAP: 19.83 -> 19.57 | No clear effect |
| EMA gamma 0.9 instead of 0.999 (ablation) | QST · Daytime->Nighttime | mAP: 20.27 -> 17.20 | Hurt |
| apply DA-BEV on top of BEVFormer | BEVFormer · Daytime->Nighttime | mAP: 15.73 -> 19.15 | Improved |
| apply DA-BEV on ResNet-50-C5 backbone | PETR · Daytime->Nighttime | mAP: 10.63 -> 13.03 | Improved |
| apply DA-BEV to 3D scene segmentation | PETR · Daytime->Nighttime segmentation | mIoU: 28.01 -> 35.59 | 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.
- MTTrans — MTTrans: Cross-Domain Object Detection with Mean Teacher Transformer (ECCV 2022) · better on 3 of 3 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
- A Black-Box Evaluation Framework for Semantic Robustness in Bird’s Eye View Detection (AAAI 2025 · similarity 0.89)
- MamBEV: Enabling State Space Models to Learn Birds-Eye-View Representations (ICLR 2025 · similarity 0.88)
- Improving Bird's Eye View Semantic Segmentation by Task Decomposition (CVPR 2024 · similarity 0.87)
- Toward Real-world BEV Perception: Depth Uncertainty Estimation via Gaussian Splatting (CVPR 2025 · similarity 0.87)
- M-BEV: Masked BEV Perception for Robust Autonomous Driving (AAAI 2024 · similarity 0.87)
- DiffBEV: Conditional Diffusion Model for Bird’s Eye View Perception (AAAI 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.
