Research map / Detection and segmentation / Open-set and few-shot detection
Unified Interaction Consistency Learning for Single-Source Domain-Generalized Object Detection in Urban Scene
Peng Zhang, Xiang Yuan, Gong Cheng
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 data augmentation to baseline detector (ablation) | Faster R-CNN · Diverse Weather, Night-Rainy | mAP: 14.5 -> 22.8 | Improved |
| train on both original and augmented images (dual pipeline) (ablation) | Faster R-CNN + DA · Diverse Weather, Night-Rainy | mAP: 22.8 -> 24.7 | Improved |
| add cross-domain interaction mechanism (CDIM) (ablation) | UICL · Diverse Weather, Night-Rainy | mAP: 24.7 -> 26.5 | Improved |
| add CIRD on top of CDIM (ablation) | UICL · Diverse Weather, Daytime-Foggy | mAP: 38.3 -> 39.4 | Improved |
| add CIRD on top of CDIM (ablation) | UICL · Diverse Weather, Night-Sunny | mAP: 44.6 -> 44.2 | Hurt |
| add CIRD on top of CDIM (ablation) | UICL · Diverse Weather, Night-Rainy | mAP: 26.5 -> 26.4 | Hurt |
| add prediction-guided consistency losses (Lba, Lbs) after CIRD (ablation) | UICL · Diverse Weather, Night-Rainy | mAP: 26.4 -> 27.4 | Improved |
| prediction-guided instead of vanilla consistency learning (ablation) | UICL · Diverse Weather, Daytime-Foggy | mAP: 39.5 -> 40.1 | Improved |
| UICL vs SECT on daytime-foggy | UICL · Diverse Weather, Daytime-Foggy | mAP: 40.6 -> 40.1 | Hurt |
Closest papers by meaning
- Dual Instance-Consistent Network for Cross-Domain Object Detection (TPAMI 2022 · similarity 0.89)
- Single-Domain Generalized Object Detection in Urban Scene via Cyclic-Disentangled Self-Distillation (CVPR 2022 · similarity 0.89)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 2024 · similarity 0.87)
- Object-Aware Domain Generalization for Object Detection (AAAI 2024 · similarity 0.86)
- CAT: Exploiting Inter-Class Dynamics for Domain Adaptive Object Detection (CVPR 2024 · similarity 0.85)
- Unbiased Faster R-CNN for Single-source Domain Generalized Object Detection (CVPR 2024 · similarity 0.85)
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.
