Research map / Detection and segmentation / Open-set and few-shot detection
Prior-based Domain Adaptive Object Detection for Hazy and Rainy Conditions
Vishwanath A. Sindagi, Poojan Oza, Rajeev Yasarla, Vishal M. Patel
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 domain adversarial alignment after conv5 (ablation) | Faster R-CNN · Cityscapes -> Foggy-Cityscapes | mAP: 24.4 -> 30.0 | Improved |
| add residual feature recovery block after conv4 (ablation) | FRCNN+D5 · Cityscapes -> Foggy-Cityscapes | mAP: 30.0 -> 32.9 | Improved |
| replace domain adversarial loss with prior-adversarial loss (ablation) | FRCNN+D5+R5 · Cityscapes -> Foggy-Cityscapes | mAP: 32.9 -> 36.5 | Improved |
| extend prior alignment and RFRB to conv4 and conv5 (ablation) | FRCNN+P5+R5 · Cityscapes -> Foggy-Cityscapes | mAP: 36.5 -> 39.3 | Improved |
| align domain at two scales without RFRB (domain loss) (ablation) | FRCNN+D5 · Cityscapes -> Foggy-Cityscapes | mAP: 30.0 -> 33.2 | Improved |
| align prior at two scales without RFRB (ablation) | FRCNN+P5+R5 · Cityscapes -> Foggy-Cityscapes | mAP: 36.5 -> 37.4 | Improved |
| use ResNet-152 backbone instead of VGG16 for full method (ablation) | FRCNN+P45+R45 · Cityscapes -> Foggy-Cityscapes | mAP: 39.3 -> 40.0 | Improved |
| add domain adversarial alignment after conv5 (ablation) | Faster R-CNN · Cityscapes -> Rainy-Cityscapes | mAP: 21.8 -> 30.2 | Improved |
| add residual feature recovery block (ablation) | FRCNN+D5 · Cityscapes -> Rainy-Cityscapes | mAP: 30.2 -> 31.2 | Improved |
| replace domain adversarial loss with prior-adversarial loss (ablation) | FRCNN+D5+R5 · Cityscapes -> Rainy-Cityscapes | mAP: 31.2 -> 33.8 | Improved |
| extend prior alignment and RFRB to conv4 and conv5 (ablation) | FRCNN+P5+R5 · Cityscapes -> Rainy-Cityscapes | mAP: 33.8 -> 36.1 | Improved |
| preprocess with Grid-Dehaze before detection | Faster R-CNN · Cityscapes -> RTTS | mAP: 30.9 -> 20.0 | Hurt |
Closest papers by meaning
- Multiple Adverse Weather Conditions Adaptation for Object Detection via Causal Intervention (TPAMI 2022 · similarity 0.89)
- Revisiting Domain-Adaptive Object Detection in Adverse Weather by the Generation and Composition of High-Quality Pseudo-Labels (ECCV 2024 · similarity 0.88)
- Image-Adaptive YOLO for Object Detection in Adverse Weather Conditions (AAAI 2022 · similarity 0.87)
- Sunshine to Rainstorm: Cross-Weather Knowledge Distillation for Robust 3D Object Detection (AAAI 2024 · similarity 0.86)
- Genuine Knowledge from Practice: Diffusion Test-Time Adaptation for Video Adverse Weather Removal (CVPR 2024 · similarity 0.85)
- Equirectangular Point Reconstruction for Domain Adaptive Multimodal 3D Object Detection in Adverse Weather Conditions (AAAI 2025 · similarity 0.84)
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.
