clovaai / Extd_pytorch

EXTD: Extremely Tiny Face Detector via Iterative Filter Reuse

A PyTorch Implementation of Extremely Tiny Face Detector via Iterative Filter Reuse

YoungJoon Yoo, Dongyoon Han, Sangdoo Yun

https://arxiv.org/abs/1906.06579

Requirement

• pytorch 1.0 (checked at 1.0)
• opencv
• numpy
• easydict
• Python3

Prepare data

WIDER face dataset is used. see the S3FD.pytorch git for more detail.

Train

You can use

python train.py 

Refer the train.py files to check the arguement. Our setting was

"--batch_size 16 --lr 0.001" 

Evaluation on WIDER Dataset

You should complie the bounding box function. Type

python3 bbox_setup.py build_ext --inplace

Then run

python3 wider_test.py

Demo

you can test your image from

python3 demo.py

References

• EXTD: Extremely Tiny Face Detector via Iterative Filter Reuse
• S3FD.pytorch

Citation

@article{yoo2019extd,
  title={EXTD: Extremely Tiny Face Detector via Iterative Filter Reuse},
  author={Yoo, YoungJoon and Han, Dongyoon and Yun, Sangdoo},
  journal={arXiv preprint arXiv:1906.06579},
  year={2019}
}

License

Copyright (c) 2019-present NAVER Corp.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE