Research map / Label-efficient and robust learning / Domain adaptation and generalization
A Free Lunch for Unsupervised Domain Adaptive Object Detection without Source Data
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 Mosaic augmentation to SFOD using SED (ablation) | SFOD · KITTI to Cityscapes | AP of Car: 43.6 -> 44.6 | Helped |
| Add Mosaic augmentation to SFOD using SED (ablation) | SFOD · Sim10k to Cityscapes | AP of Car: 42.3 -> 42.9 | Helped |
| Add Mosaic augmentation to SFOD using SED (ablation) | SFOD · Cityscapes to BDD100k daytime | mAP: 27.6 -> 29.0 | Helped |
| Add Mosaic augmentation to SFOD using SED (ablation) | SFOD · Cityscapes to Foggy Cityscapes | mAP: 30.6 -> 33.5 | Helped |
| Use SED-selected rather than ideal confidence threshold (ablation) | SFOD · KITTI to Cityscapes | AP of Car: 43.7 -> 43.6 | No clear effect |
| Use SED-selected rather than ideal confidence threshold (ablation) | SFOD · Sim10k to Cityscapes | AP of Car: 42.5 -> 42.3 | No clear effect |
| Use SED-selected rather than ideal confidence threshold (ablation) | SFOD · Cityscapes to BDD100k daytime | mAP: 27.7 -> 27.6 | No clear effect |
| Use SED-selected rather than ideal confidence threshold (ablation) | SFOD · Cityscapes to Foggy Cityscapes | mAP: 31.0 -> 30.6 | Hurt |
| Add target-image defogging to SFOD using SED (ablation) | SFOD · Cityscapes to Foggy Cityscapes | mAP: 30.6 -> 34.2 | Helped |
| Add target-image defogging to SFOD-Mosaic using SED (ablation) | SFOD-Mosaic · Cityscapes to Foggy Cityscapes | mAP: 33.5 -> 36.3 | Helped |
Closest papers by meaning
- Pseudo Label Refinery for Unsupervised Domain Adaptation on Cross-dataset 3D Object Detection (CVPR 2024 · similarity 0.89)
- Active Domain Adaptation with False Negative Prediction for Object Detection (CVPR 2024 · similarity 0.89)
- Large Self-Supervised Models Bridge the Gap in Domain Adaptive Object Detection (CVPR 2025 · similarity 0.89)
- Towards Unsupervised Model Selection for Domain Adaptive Object Detection (NeurIPS 2024 · similarity 0.88)
- SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection (CVPR 2025 · similarity 0.88)
- Universal Domain Adaptive Object Detection via Dual Probabilistic Alignment (AAAI 2025 · similarity 0.87)
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.