PyTorch implementation of "SWDet: Anchor-based Object Detector for Solid Waste Detection in Aerial Images", [IEEE JSTARS, 2022].
Method | Backbone | AP50 | AP75 | Time | Size | Config | Download |
---|---|---|---|---|---|---|---|
ATSS | R-50-FPN | 72.74 | 50.07 | 50.4 | 31.90M | config | model |
AutoAssign | R-50-FPN | 75.67 | 50.16 | 49.7 | 35.98M | config | model |
Fovea | R-50-FPN | 74.05 | 49.39 | 48.3 | 36.02M | config | model |
PAA | R-50-FPN | 73.55 | 48.93 | 74.0 | 31.90M | config | model |
VFNet | R-50-FPN | 71.08 | 47.86 | 55.6 | 32.49M | config | model |
YOLOF | R-50-FPN | 76.57 | 47.64 | 32.7 | 42.16M | config | model |
YOLOv5s | CSPDarknet | 74.33 | 53.27 | 2.1 | 7.02M | config | model |
SWDet | R-50-EAFPN | 74.29 | 55.08 | 5.0 | 17.90M | config | model |
SWDet | ADA-EAFPN | 77.58 | 58.39 | 9.4 | 33.85M | config | model |
@article{swdet,
author={Zhou, Liming and Rao, Xiaohan and Li, Yahui and Zuo, Xianyu and Liu, Yang and Lin, Yinghao and Yang, Yong},
journal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
title={SWDet: Anchor-based Object Detector for Solid Waste Detection in Aerial Images},
year={2022},
pages={1-15},
doi={10.1109/JSTARS.2022.3218958}
}
Any question regarding this work can be addressed to [email protected].