In this repo, you can find the weight file created by training with YOLOv3 and our results on the WIDER dataset. Also it has been added configuration files for use of weights file properly.
If you want to test our system you can use the uploaded sample code. For using it you must install darknet and pydarknet.
If you use this software and weight in your research, please cite our publication:
@INPROCEEDINGS{Yolov3face_2019_eleco,
author={Gurkan, Filiz and Sagman, Bunyamin and Gunsel, Bilge},
booktitle={2019 11th International Conference on Electrical and Electronics Engineering (ELECO)},
title={YOLOv3 as a Deep Face Detector},
year={2019},
volume={},
number={},
pages={605-609},
doi={10.23919/ELECO47770.2019.8990641}}