Research map / Detection and segmentation / Open-set and few-shot detection
Domain Adaptive Object Detection via Balancing Between Self-Training and Adversarial Learning
Muhammad Akhtar Munir, Muhammad Haris Khan, Muhammad Saquib Sarfraz, Mohsen Ali
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 uncertainty-guided pseudo-labelling (UGPL+) (ablation) | FCOS baseline · Sim10k to Cityscapes | AP@0.5: 46.0 -> 50.2 | Improved |
| add uncertainty-guided tiling (UGT+) (ablation) | FCOS baseline · Sim10k to Cityscapes | AP@0.5: 46.0 -> 51.5 | Improved |
| combine UGPL+ and UGT+ (SSAL+) (ablation) | UGPL+ · Sim10k to Cityscapes | AP@0.5: 50.2 -> 53.0 | Improved |
| use confidence-based pseudo-labels instead of uncertainty-guided (ablation) | Baseline · Sim10k to Cityscapes | AP@0.5: 46.0 -> 43.2 | Hurt |
| add localization condition to pseudo-label selection (ablation) | SSAL · Sim10k to Cityscapes | AP@0.5: 48.2 -> 51.8 | Improved |
| add localization condition to pseudo-label selection (ablation) | SSAL+ · Sim10k to Cityscapes | AP@0.5: 50.3 -> 53.0 | Improved |
| tile around uncertain detections (UGT+) instead of random tiles (ablation) | UGPL+ · Sim10k to Cityscapes | AP@0.5: 51.1 -> 53.0 | Improved |
| tile around uncertain instead of certain detections (ablation) | UGPL+ · Sim10k to Cityscapes | AP@0.5: 52.2 -> 53.0 | Improved |
| increase MC-dropout forward passes from 10 to 20 or 30 (ablation) | SSAL+ · Sim10k to Cityscapes | AP@0.5: 53.0 -> 52.4 | No clear effect |
| add rounds R1 and R2 (self-training plus UGT+) after R0 (ablation) | SSAL+ · Sim10k to Cityscapes | AP@0.5: 47.8 -> 53.0 | Improved |
| add revisited variance constraint and new tile set (SSAL+ over SSAL) | SSAL · Sim10k to Cityscapes | AP@0.5: 51.8 -> 53.0 | Improved |
| add SSAL+ over SSAL | SSAL · Cityscapes to Foggy Cityscapes | mAP@0.5: 39.6 -> 39.8 | No clear effect |
Closest papers by meaning
- SSAL: Synergizing between Self-Training and Adversarial Learning for Domain Adaptive Object Detection (NeurIPS 2021 · similarity 0.97)
- Universal Domain Adaptive Object Detection via Dual Probabilistic Alignment (AAAI 2025 · similarity 0.89)
- Differential Alignment for Domain Adaptive Object Detection (AAAI 2025 · similarity 0.88)
- Large Self-Supervised Models Bridge the Gap in Domain Adaptive Object Detection (CVPR 2025 · similarity 0.88)
- Toward Improving Robustness and Accuracy in Unsupervised Domain Adaptation (AAAI 2025 · similarity 0.88)
- Mean Teacher DETR with Masked Feature Alignment: A Robust Domain Adaptive Detection Transformer Framework (AAAI 2024 · similarity 0.88)
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.
