Research map / Detection and segmentation / Open-set and few-shot detection
RPN Prototype Alignment for Domain Adaptive Object Detector
Check 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 RPN prototype alignment using IoU-assigned pseudo-label proposals (ablation) | Faster R-CNN, VGG16 · Cityscapes → Foggy Cityscapes | mAP: 37.0 -> 38.5 | Helped |
| add Grad CAM spatial weighting to RPN prototype alignment (ablation) | RPN prototype alignment · Cityscapes → Foggy Cityscapes | mAP: 38.5 -> 39.0 | Helped |
| add RPN prototype alignment using IoU-assigned pseudo-label proposals (ablation) | Faster R-CNN, VGG16 · SIM10k → Cityscapes | car AP: 42.3 -> 45.3 | Helped |
| add Grad CAM spatial weighting to RPN prototype alignment (ablation) | RPN prototype alignment · SIM10k → Cityscapes | car AP: 45.3 -> 45.7 | Helped |
| add RPN prototype alignment using IoU-assigned pseudo-label proposals (ablation) | Faster R-CNN, VGG16 · Cityscapes → KITTI | mAP: 41.6 -> 44.2 | Helped |
| add Grad CAM spatial weighting to RPN prototype alignment (ablation) | RPN prototype alignment · Cityscapes → KITTI | mAP: 44.2 -> 44.8 | Helped |
| replace IoU-based proposal labels with hard RPC probability labels (ablation) | RPN prototype alignment · Cityscapes → Foggy Cityscapes | mAP: 39.0 -> 37.5 | Hurt |
| replace IoU-based proposal labels with soft RPC probability labels (ablation) | RPN prototype alignment · Cityscapes → Foggy Cityscapes | mAP: 39.0 -> 37.8 | Hurt |
| use pseudo labels to train RPN classification only (ablation) | Faster R-CNN RPN · Cityscapes → Foggy Cityscapes | mAP: 39.0 -> 37.3 | Hurt |
| also apply pseudo labels to RPN box regression (ablation) | Faster R-CNN RPN · SIM10k → Cityscapes | mAP: 42.3 -> 39.2 | Hurt |
| vary the class-agnostic pseudo-label retention ratio (ablation) | RPN prototype alignment · Cityscapes → Foggy Cityscapes | mAP: 37.8; 38.2; 38.7; 39.0; 38.9; 38.5 | Mixed |
| use our RPN proposals with HTCN at inference | HTCN · Cityscapes → Foggy Cityscapes | mAP: 39.8 -> 40.5 | Helped |
Closest papers by meaning
- Cross-Domain Detection via Graph-Induced Prototype Alignment (CVPR 2020 · similarity 0.91)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 2024 · similarity 0.90)
- Universal Domain Adaptive Object Detection via Dual Probabilistic Alignment (AAAI 2025 · similarity 0.89)
- Differential Alignment for Domain Adaptive Object Detection (AAAI 2025 · similarity 0.87)
- Mean Teacher DETR with Masked Feature Alignment: A Robust Domain Adaptive Detection Transformer Framework (AAAI 2024 · similarity 0.87)
- Large Self-Supervised Models Bridge the Gap in Domain Adaptive Object Detection (CVPR 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.