Research map / Detection and segmentation / Open-set and few-shot detection
Every Pixel Matters: Center-aware Feature Alignment for Domain Adaptive Object Detector
Cheng-Chun Hsu, Yi-Hsuan Tsai, Yen-Yu Lin, Ming-Hsuan Yang
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 center-aware alignment on top of global alignment (ablation) | Ours (GA), VGG-16 · Cityscapes -> Foggy Cityscapes | mAP@0.5: 33.2 -> 36.0 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), ResNet-101 · Cityscapes -> Foggy Cityscapes | mAP@0.5: 36.6 -> 40.2 | Improved |
| replace global alignment with center-aware alignment (ablation) | Ours, VGG-16 · Cityscapes -> Foggy Cityscapes | mAP@0.5: 33.2 -> 34.2 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), VGG-16 · Sim10k -> Cityscapes | mAP@0.5: 45.9 -> 49.0 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), ResNet-101 · Sim10k -> Cityscapes | mAP@0.5: 50.6 -> 51.2 | No clear effect |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), ResNet-101 · Sim10k -> Cityscapes | mAP@0.75: 25.2 -> 27.4 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), ResNet-101 · Sim10k -> Cityscapes | mAP medium objects: 26.3 -> 30.2 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), VGG-16 · KITTI -> Cityscapes | mAP@0.5: 39.1 -> 43.2 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), ResNet-101 · KITTI -> Cityscapes | mAP@0.5: 42.3 -> 45.0 | Improved |
| add center-aware alignment on top of global alignment (ablation) | Ours (GA), ResNet-101 · KITTI -> Cityscapes | mAP: 18.8 -> 22.2 | Improved |
| add domain adaptation (GA+CA) to the detector | Ours (w/o adapt.), VGG-16 · Cityscapes -> Foggy Cityscapes | mAP@0.5: 18.4 -> 36.0 | Improved |
| add domain adaptation (GA+CA) to the detector | Ours (w/o adapt.), VGG-16 · KITTI -> Cityscapes | mAP@0.5: 34.4 -> 43.2 | Improved |
Closest papers by meaning
- Adapting Object Detectors via Selective Cross-Domain Alignment (CVPR 2019 · similarity 0.91)
- Universal Domain Adaptive Object Detection via Dual Probabilistic Alignment (AAAI 2025 · similarity 0.89)
- Robust Domain Adaptive Object Detection With Unified Multi-Granularity Alignment (TPAMI 2024 · similarity 0.89)
- Multi-Granularity Alignment Domain Adaptation for Object Detection (CVPR 2022 · similarity 0.89)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 2024 · similarity 0.88)
- Differential Alignment for Domain Adaptive Object Detection (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.
