DRAFE: Domain-Robust Asymmetric Fusion of Heterogeneous Detection Transformers for Cross-City Fine-Grained Traffic Object Detection
2026-08-17 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed a system called DRAFE to better detect and recognize different types of vehicles across cities. They combined two separate detectors trained on a large, carefully labeled dataset, then fine-tuned this combination on a specific challenge dataset. Their method includes smart ways to merge the detectors' outputs to improve accuracy. In a vehicle detection competition, their approach ranked 6th out of 25 teams and showed noticeable improvement over an earlier version. This work helps make traffic monitoring systems more reliable in different city environments.
deep learningobject detectiondomain generalizationvehicle recognitionensemble methodspseudo-labelingdataset annotationmean Average Precision (mAP)anchor-based detectionconfidence recalibration
Authors
Divine Yao Agbobli, Geoffery Eyram Agorku, Israel Afriyie, Kwadwo Amankwah-Nkyi, Marvin Osei-Kuffour, Richmond Owusu Duah, Bright Seglah, Kelvin Asamoah Terkper, Kwabena Amoako Adjei
Abstract
Deep learning-based object detectors are fundamental to intelligent transportation systems, enabling traffic monitoring, vehicle analytics, and infrastructure management. However, achieving both fine-grained vehicle recognition and robust cross-city domain generalization remains challenging. We present the Domain-Robust Asymmetric Fusion Ensemble (DRAFE), which combines independently trained LW-DETR and RF-DETR detectors for cross-city fine-grained traffic object detection. DRAFE employs a two-stage training strategy that first pretrains complementary detectors on diverse public traffic datasets using pseudo-label expansion and human-in-the-loop annotation refinement, producing a curated corpus of 6,049 images and 203,619 annotations, before challenge-compliant fine-tuning on the Project Hafnia Track 6 dataset. At inference, DRAFE applies anchor-conditioned class-consistent matching, reliability-weighted coordinate fusion, agreement-aware confidence recalibration, and complementary hypothesis recovery. On AI City Challenge 2026 Track 6, DRAFE achieves 0.4022 mAP, ranks sixth among 25 participating teams, and improves by 0.0553 mAP over a preliminary ensemble evaluated under identical benchmark conditions.