DG-DETR: Toward Domain Generalized Detection Transformer
Not on the map. Why?
Abstract (arXiv)
End-to-end Transformer-based detectors (DETRs) have demonstrated strong detection performance. However, domain generalization (DG) research has primarily focused on convolutional neural network (CNN)-based detectors, while paying little attention to enhancing the robustness of DETRs. In this letter, we introduce a Domain Generalized DEtection TRansformer (DG-DETR), a simple, effective, and plug-and-play method that improves out-of-distribution (OOD) robustness for DETRs. Specifically, we propose a novel domain-agnostic query selection strategy that removes domain-induced biases from object queries via orthogonal projection onto the instance-specific style space. Additionally, we leverage a wavelet decomposition to disentangle features into domain-invariant and domain-specific components, enabling synthesis of diverse latent styles while preserving the semantic features of objects. Experimental results validate the effectiveness of DG-DETR. Our code is available at https://github.com/sminhwang/DG-DETR.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 48 of its 54 references that are in the map. First posted on arXiv 28 Apr 2025.
Builds on
48 of 54 references.
References · 48
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
- 2020End-to-End Object Detection with Transformerscited by 7,414
- 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
- 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
- 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
- 2020Supervised Contrastive Learningcited by 2,829
- 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,577
- 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
- 2022DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detectioncited by 1,319
- 2018BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learningcited by 1,288
- 2021Domain Generalization: A Surveycited by 764
- 2022DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETRcited by 573
36 more
- 2021Domain Generalization with MixStylecited by 566
- 2023DETRs Beat YOLOs on Real-time Object Detectioncited by 533
- 2022DN-DETR: Accelerate DETR Training by Introducing Query DeNoisingcited by 488
- 2021Intriguing Properties of Vision Transformerscited by 446
- 2018Domain Adaptive Faster R-CNN for Object Detection in the Wildcited by 443
- 2019Domain Generalization via Model-Agnostic Learning of Semantic Featurescited by 407
- 2017Semantic Foggy Scene Understanding with Synthetic Datacited by 403
- 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 364
- 2019Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Comingcited by 336
- 2018Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptationcited by 218
- 2021RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whiteningcited by 215
- 2021Learning to Diversify for Single Domain Generalizationcited by 170
- 2021FSCE: Few-Shot Object Detection via Contrastive Proposal Encodingcited by 168
- 2019Photorealistic Style Transfer via Wavelet Transformscited by 166
- 2020Unbiased Mean Teacher for Cross-domain Object Detectioncited by 159
- 2022DETRs with Hybrid Matchingcited by 150
- 2019Progressive Domain Adaptation for Object Detectioncited by 134
- 2020Exploring Categorical Regularization for Domain Adaptive Object Detectioncited by 131
- 2021Cross-Domain Adaptive Teacher for Object Detectioncited by 126
- 2022Uncertainty Modeling for Out-of-Distribution Generalizationcited by 125
- 2021Open Domain Generalization with Domain-Augmented Meta-Learningcited by 115
- 2019Switchable Whitening for Deep Representation Learningcited by 106
- 2019Iterative Normalization: Beyond Standardization towards Efficient Whiteningcited by 92
- 2023CLIP the Gap: A Single Domain Generalization Approach for Object Detectioncited by 87
- 2021Adversarially Adaptive Normalization for Single Domain Generalizationcited by 85
- 2021Delving Deep into the Generalization of Vision Transformers under Distribution Shiftscited by 85
- 2022Improving Vision Transformers by Revisiting High-frequency Componentscited by 76
- 2020Informative Dropout for Robust Representation Learning: A Shape-bias Perspectivecited by 57
- 2023Contrastive Mean Teacher for Domain Adaptive Object Detectorscited by 53
- 2021Feature Stylization and Domain-aware Contrastive Learning for Domain Generalizationcited by 40
- 2024Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignmentcited by 29
- 2023ALOFT: A Lightweight MLP-like Architecture with Dynamic Low-frequency Transform for Domain Generalizationcited by 28
- 2021Robust Object Detection via Instance-Level Temporal Cycle Confusioncited by 24
- 2023Object-Aware Domain Generalization for Object Detectioncited by 21
- 2024G-NAS: Generalizable Neural Architecture Search for Single Domain Generalization Object Detectioncited by 17
- 2024PhysAug: A Physical-guided and Frequency-based Data Augmentation for Single-Domain Generalized Object Detectioncited by 11
Built on by
Citing papers per year: 2026: 3.
3 papers.