FSDETR: Frequency-Spatial Feature Enhancement for Small Object Detection

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Huang, Jianchao, Zhang, Fengming, Zhu, Haibo, Yan, Tao
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908970281598976
author Huang, Jianchao
Zhang, Fengming
Zhu, Haibo
Yan, Tao
author_facet Huang, Jianchao
Zhang, Fengming
Zhu, Haibo
Yan, Tao
contents Small object detection remains a significant challenge due to feature degradation from downsampling, mutual occlusion in dense clusters, and complex background interference. To address these issues, this paper proposes FSDETR, a frequency-spatial feature enhancement framework built upon the RT-DETR baseline. By establishing a collaborative modeling mechanism, the method effectively leverages complementary structural information. Specifically, a Spatial Hierarchical Attention Block (SHAB) captures both local details and global dependencies to strengthen semantic representation. Furthermore, to mitigate occlusion in dense scenes, the Deformable Attention-based Intra-scale Feature Interaction (DA-AIFI) focuses on informative regions via dynamic sampling. Finally, the Frequency-Spatial Feature Pyramid Network (FSFPN) integrates frequency filtering with spatial edge extraction via the Cross-domain Frequency-Spatial Block (CFSB) to preserve fine-grained details. Experimental results show that with only 14.7M parameters, FSDETR achieves 13.9% APS on VisDrone 2019 and 48.95% AP50 tiny on TinyPerson, showing strong performance on small-object benchmarks. The code and models are available at https://github.com/YT3DVision/FSDETR.
format Preprint
id arxiv_https___arxiv_org_abs_2604_14884
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FSDETR: Frequency-Spatial Feature Enhancement for Small Object Detection
Huang, Jianchao
Zhang, Fengming
Zhu, Haibo
Yan, Tao
Computer Vision and Pattern Recognition
Small object detection remains a significant challenge due to feature degradation from downsampling, mutual occlusion in dense clusters, and complex background interference. To address these issues, this paper proposes FSDETR, a frequency-spatial feature enhancement framework built upon the RT-DETR baseline. By establishing a collaborative modeling mechanism, the method effectively leverages complementary structural information. Specifically, a Spatial Hierarchical Attention Block (SHAB) captures both local details and global dependencies to strengthen semantic representation. Furthermore, to mitigate occlusion in dense scenes, the Deformable Attention-based Intra-scale Feature Interaction (DA-AIFI) focuses on informative regions via dynamic sampling. Finally, the Frequency-Spatial Feature Pyramid Network (FSFPN) integrates frequency filtering with spatial edge extraction via the Cross-domain Frequency-Spatial Block (CFSB) to preserve fine-grained details. Experimental results show that with only 14.7M parameters, FSDETR achieves 13.9% APS on VisDrone 2019 and 48.95% AP50 tiny on TinyPerson, showing strong performance on small-object benchmarks. The code and models are available at https://github.com/YT3DVision/FSDETR.
title FSDETR: Frequency-Spatial Feature Enhancement for Small Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.14884