ECAN-Detector: An Efficient Context-Aggregation Network for Small-Object Detection
AppliedMath · 20.05.2025 · 5 (2) · 58
DOI: 10.3390/appliedmath5020058 ↗Kurz gefasst
Fachartikel in der Zeitschrift AppliedMath, online erschienen am 20.05.2025. Verfasst von Gaofeng Xing und 5 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Shaanxi Province Qinchuangyuan “Scientist + Engineer” Team Construction Project.
Abstract (Originalsprache)
Over the past decade, the field of object detection has advanced remarkably, especially in the accurate recognition of medium- and large-sized objects. Nevertheless, detecting small objects is still difficult because their low-resolution appearance provides insufficient discriminative features, and they often suffer severe occlusions, particularly in the safety-critical context of autonomous driving. Conventional detectors often fail to extract sufficient information from shallow feature maps, which limits their ability to detect small objects with high precision. To address this issue, we propose the ECAN-Detector, an efficient context-aggregation method designed to enrich the feature representation of shallow layers, which are particularly beneficial for small-object detection. The model first employs an additional shallow detection layer to extract high-resolution features that provide more detailed information for subsequent stages of the network, and then incorporates a dynamic scaled transformer (DST) that enriches spatial perception by adaptively fusing global semantics and local context. Concurrently, a context-augmentation module (CAM) embedded in the shallow layer complements both global and local features relevant to small objects. To further boost the average precision of small-object detection, we implement a faster method utilizing two reparametrized convolutions in the detection head. Finally, extensive experiments conducted on the VisDrone2012-DET and VisDrone2021-DET datasets verified that our proposed method surpasses the baseline model, and achieved a significant improvement of 3.1% in AP and 3.5% in APs. Compared with recent state-of-the-art (SOTA) detectors, ECAN Detector delivers comparable accuracy yet preserves real-time throughput, reaching 54.3 FPS.
Förderung
- National Natural Science Foundation of China
- Shaanxi Province Qinchuangyuan “Scientist + Engineer” Team Construction Project
Zitieren
APA
Xing, G., Xu, Z., He, Y., Ning, H., Sun, M., & Wang, C. (2025). ECAN-Detector: An Efficient Context-Aggregation Network for Small-Object Detection. AppliedMath, 5(2), 58. https://doi.org/10.3390/appliedmath5020058
Vancouver
Xing G, Xu Z, He Y, Ning H, Sun M, Wang C. ECAN-Detector: An Efficient Context-Aggregation Network for Small-Object Detection. AppliedMath. 2025;5(2):58. doi:10.3390/appliedmath5020058
BibTeX
@article{Xing2025,
author = {Xing, Gaofeng and Xu, Zhikang and He, Yulong and Ning, Hailong and Sun, Menghao and Wang, Chunmei},
title = {ECAN-Detector: An Efficient Context-Aggregation Network for Small-Object Detection},
journal = {AppliedMath},
year = {2025},
volume = {5},
number = {2},
pages = {58},
doi = {10.3390/appliedmath5020058}
}
EUVIMED