BlazeFace_person_pytorch

Blazeface trained on pascal_voc person.

This repo is an unofficial implementation of:

BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs Paper

The repo contains..

The training and inference is > 10x faster than SSD.

The BlazeFace model is based on here, and localaization and detection layers are added from ssd.pytorch.

Also the nms functions are borrowed from ssd.pytorch as well.

Thank you!

requirements

Pytorch > 1.0
opencv
scikit-learn

Run:

python train_BlazeFace.py

See train_BlazeFace.ipynb for specifics.

Dataset_test_with_BlazeFace128-VOC-allclasses.ipynb runs all VOC classes, but doesn't get good accuracy at this point.

See inference.ipynb to run inference.

It takes less than 30ms on K80 to run single image (including nms).

Name		Name	Last commit message	Last commit date
Latest commit History 45 Commits
log		log
old		old
sample		sample
utils		utils
Dataset_test_with_BlazeFace.ipynb		Dataset_test_with_BlazeFace.ipynb
Dataset_test_with_BlazeFace128-VOC-allclasses.ipynb		Dataset_test_with_BlazeFace128-VOC-allclasses.ipynb
Dataset_test_with_BlazeFace256.ipynb		Dataset_test_with_BlazeFace256.ipynb
Dataset_test_with_SSD.ipynb		Dataset_test_with_SSD.ipynb
LICENSE		LICENSE
README.md		README.md
eval.ipynb		eval.ipynb
inference-256.ipynb		inference-256.ipynb
inference.ipynb		inference.ipynb
train_BlazeFace-Copy1.ipynb		train_BlazeFace-Copy1.ipynb
train_BlazeFace.ipynb		train_BlazeFace.ipynb
train_BlazeFace.py		train_BlazeFace.py
train_BlazeFace256.ipynb		train_BlazeFace256.ipynb