Salvato in:
Dettagli Bibliografici
Autori principali: Pan, Weichao, Kang, Jiaju, Wang, Xu, Chen, Zhihao, Ge, Yiyuan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2409.01604
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912013067747328
author Pan, Weichao
Kang, Jiaju
Wang, Xu
Chen, Zhihao
Ge, Yiyuan
author_facet Pan, Weichao
Kang, Jiaju
Wang, Xu
Chen, Zhihao
Ge, Yiyuan
contents Current road damage detection methods, relying on manual inspections or sensor-mounted vehicles, are inefficient, limited in coverage, and often inaccurate, especially for minor damages, leading to delays and safety hazards. To address these issues and enhance real-time road damage detection using street view image data (SVRDD), we propose DAPONet, a model incorporating three key modules: a dual attention mechanism combining global and local attention, a multi-scale partial over-parameterization module, and an efficient downsampling module. DAPONet achieves a mAP50 of 70.1% on the SVRDD dataset, outperforming YOLOv10n by 10.4%, while reducing parameters to 1.6M and FLOPs to 1.7G, representing reductions of 41% and 80%, respectively. On the MS COCO2017 val dataset, DAPONet achieves an mAP50-95 of 33.4%, 0.8% higher than EfficientDet-D1, with a 74% reduction in both parameters and FLOPs.
format Preprint
id arxiv_https___arxiv_org_abs_2409_01604
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DAPONet: A Dual Attention and Partially Overparameterized Network for Real-Time Road Damage Detection
Pan, Weichao
Kang, Jiaju
Wang, Xu
Chen, Zhihao
Ge, Yiyuan
Computer Vision and Pattern Recognition
Current road damage detection methods, relying on manual inspections or sensor-mounted vehicles, are inefficient, limited in coverage, and often inaccurate, especially for minor damages, leading to delays and safety hazards. To address these issues and enhance real-time road damage detection using street view image data (SVRDD), we propose DAPONet, a model incorporating three key modules: a dual attention mechanism combining global and local attention, a multi-scale partial over-parameterization module, and an efficient downsampling module. DAPONet achieves a mAP50 of 70.1% on the SVRDD dataset, outperforming YOLOv10n by 10.4%, while reducing parameters to 1.6M and FLOPs to 1.7G, representing reductions of 41% and 80%, respectively. On the MS COCO2017 val dataset, DAPONet achieves an mAP50-95 of 33.4%, 0.8% higher than EfficientDet-D1, with a 74% reduction in both parameters and FLOPs.
title DAPONet: A Dual Attention and Partially Overparameterized Network for Real-Time Road Damage Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.01604