Research map / Detection and segmentation / Open-set and few-shot detection
Domain-adaptive object detection
Adapt detectors across domains with alignment, diversity, and mean-teacher methods.
103 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2018–2026) · an approach in Open-set and few-shot detection, keyword cluster “domain · adaptation · adaptive object”
State History Benchmarks Tried Baselines Checklist Preprints Who Export
Where it stands
Still moving The best reported result is still improving on 1 of 1 tracked benchmark.
29 papers in 2023–24 and 41 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 4.9% to 7.4%.
Open questions the data points to
- Cityscapes → Foggy Cityscapes (mAP): best went from 52.3 (up to 2024) to 57.5 in 2025 — SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection (CVPR 2025 · 11 citations)
- Papers report DA-Faster on Cityscapes→FoggyCityscapes (mAP) anywhere from 27.6 to 39.6: comparisons against it depend on whose number you take.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2018 · Domain Adaptive Faster R-CNN for Object Detection in the Wild (CVPR 2018 · 1,649 citations)
- 2019 · Strong-Weak Distribution Alignment for Adaptive Object Detection (CVPR 2019 · 824 citations)
- 2020 · Exploring Categorical Regularization for Domain Adaptive Object Detection (CVPR 2020 · 337 citations)
- 2021 · Unbiased Mean Teacher for Cross-Domain Object Detection (CVPR 2021 · 374 citations)
- 2022 · Single-Domain Generalized Object Detection in Urban Scene via Cyclic-Disentangled Self-Distillation (CVPR 2022 · 129 citations)
- 2023 · CLIP the Gap: A Single Domain Generalization Approach for Object Detection (CVPR 2023 · 127 citations)
- 2024 · Boosting Object Detection with Zero-Shot Day-Night Domain Adaptation (CVPR 2024 · 71 citations)
- 2025 · PhysAug: A Physical-guided and Frequency-based Data Augmentation for Single-Domain Generalized Object Detection (AAAI 2025 · 13 citations)
- 2026 · Cross-domain Joint Learning with Prototype-guided Mixture-of-Experts for Infrared Moving Small Target Detection (AAAI 2026 · 2 citations)
Earliest papers here
- Domain Adaptive Faster R-CNN for Object Detection in the Wild (CVPR 2018 · 1,649 citations)
- LSTD: A Low-Shot Transfer Detector for Object Detection (AAAI 2018 · 341 citations)
- People, Penguins and Petri Dishes: Adapting Object Counting Models to New Visual Domains and Object Types Without Forgetting (CVPR 2018 · 91 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
Cityscapes → Foggy Cityscapes · mAP (higher is better)
8 papers, 2020–2025 · still improving
| Year | Best so far | Set by |
|---|---|---|
| 2020 | 36.2 | iFAN: Image-Instance Full Alignment Networks for Adaptive Object Detection (AAAI 2020 · 83 citations) |
| 2021 | 41.8 | MeGA-CDA: Memory Guided Attention for Category-Aware Unsupervised Domain Adaptive Object Detection (CVPR 2021 · 214 citations) |
| 2024 | 52.3 | DSD-DA: Distillation-based Source Debiasing for Domain Adaptive Object Detection (ICML 2024 · 1 citation) |
| 2025 | 57.5 | SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection (CVPR 2025 · 11 citations) |
What has been tried
73 papers here report 871 measured changes (a component added, removed or swapped, compared with a baseline): helped 82%, no clear effect 3%, hurt 14%, mixed 1%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Train directly on synthesized low-light images | WIDER FACE→DARK FACE · mAP(%) | 15.2% -> 11.6% | Hurt | Boosting Object Detection with Zero-Shot Day-Night Domain Adaptation (CVPR 2024 · 71 citations) |
| Add single-level transformation without mixing | Cityscapes-C · mPC | 15.4 -> 16.8 | Helped | Object-Aware Domain Generalization for Object Detection (AAAI 2024 · 47 citations) |
| Add Global-Local Transformation | Daytime-Clear · mAP@0.5 | 56.2 -> 60.4 | Helped | Unbiased Faster R-CNN for Single-source Domain Generalized Object Detection (CVPR 2024 · 46 citations) |
| Add WST features to SAR input | SARDet-100K test · mAP | 50.2 -> 51.1 | Helped | SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection (NeurIPS 2024 · 33 citations) |
| Add diversification to the Faster R-CNN baseline | Pascal VOC → Clipart1k · mAP@0.5 (%) | 25.7 -> 34.2 | Helped | Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 2024 · 24 citations) |
| add NAS | Four target domains average · mAP | +2.4% mAP with G-loss; +1.6% mAP without G-loss | Helped | G-NAS: Generalizable Neural Architecture Search for Single Domain Generalization Object Detection (AAAI 2024 · 23 citations) |
| add Dual-Teachers | RGB→Thermal FLIR · mAP | 65.81 -> 66.93 | Helped | D3T: Distinctive Dual-Domain Teacher Zigzagging Across RGB-Thermal Gap for Domain-Adaptive Object Detection (CVPR 2024 · 18 citations) |
| Replace non-learnable daytime augmentation with self-supervised ISP degradation | BDD100k night val · AP | 42.2 -> 48.5 | Helped | ISP-Teacher:Image Signal Process with Disentanglement Regularization for Unsupervised Domain Adaptive Dark Object Detection (AAAI 2024 · 17 citations) |
| Add antipodal regularization | GraspNet-1billion RealSense novel · AP | 32.38 -> 35.58 | Helped | Generalizing 6-DoF Grasp Detection via Domain Prior Knowledge (CVPR 2024 · 14 citations) |
| add non-uniform illumination to local occlusion | DWD · mPC | 35.2 -> 37.5 | Helped | PhysAug: A Physical-guided and Frequency-based Data Augmentation for Single-Domain Generalized Object Detection (AAAI 2025 · 13 citations) |
| replace self-attention with domain-aware attention | Cross-Weather, Cross-FoV, Sim-to-Real, and Cross-Style · mAP | 54.8 -> 57.5; 61.9 -> 67.1; 62.3 -> 66.8; 43.1 -> 47.9 | Helped | SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection (CVPR 2025 · 11 citations) |
| Add MIC strong augmentation | Cityscapes → Foggy Cityscapes · AP50 | 50.0 -> 53.9 | Helped | Differential Alignment for Domain Adaptive Object Detection (AAAI 2025 · 9 citations) |
Baseline numbers that disagree
The same baseline on the same benchmark and metric, reported more than 5% apart by different papers here. Training budgets, data splits and reimplementations differ, so state which number you compare with.
| Baseline | Benchmark | Metric | Reported | Papers |
|---|---|---|---|---|
| DA-Faster | Cityscapes→FoggyCityscapes | mAP | 27.6 – 39.6 | 2 papers, e.g. AAAI 2020 and ICML 2024 |
What reviewers will expect
Datasets and baselines named in the 15 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- Cityscapes · 6 of 15 papers
- Foggy Cityscapes · 4 of 15 papers
- Sim10k · 3 of 15 papers
- Clipart1k · 3 of 15 papers
- Cityscapes-C · 2 of 15 papers
- Pascal VOC · 2 of 15 papers
- Watercolor · 2 of 15 papers
- Clipart · 2 of 15 papers
Baselines
- SIGMA · 4 of 15 papers
- Faster R-CNN · 4 of 15 papers
- AT · 3 of 15 papers
- UFR · 3 of 15 papers
- HT · 2 of 15 papers
- SFA · 2 of 15 papers
- DA-Faster · 2 of 15 papers
- Source Only · 2 of 15 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 70 papers; from OpenAlex and the papers' affiliations).
- Hong Kong University of Science and Technology · 3 papers
- University of Chinese Academy of Sciences · 3 papers
- Chinese Academy of Sciences · 3 papers
- Beihang University · 3 papers
- Nankai University · 3 papers
- Tsinghua University · 2 papers
- Huawei · 2 papers
- Harbin Institute of Technology · 2 papers
- East China Normal University · 2 papers
- The Chinese University of Hong Kong · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Domain-adaptive segmentation (27 papers)
- Cross-domain translation and pose (30 papers)
Elsewhere in the topic
- Object-level representation learning (219 papers)
- Interactive and referring segmentation (161 papers)
- Open-set and unknown object detection (160 papers)
- Amodal and open-world instance segmentation (137 papers)
- Industrial anomaly detection (117 papers)
- Unsupervised object and part discovery (114 papers)
- Open-vocabulary object detection (96 papers)
- Open-vocabulary semantic segmentation (93 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).