faster_rcnn_pytorch

clone repo

git clone https://github.com/jerinka/FastRCNN_Pytorch.git -b sign Sign_Det

Introduction

This is a PyTorch project using Faster RCNN for 2-class face mask detection.

For Faster RCNN tutorial, please see: https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html

Dataset Description

Kaggle face mask detection dataset: https://www.kaggle.com/andrewmvd/face-mask-detection

• contains 853 images
• each image is accompanied by an annotation file, including multiple bounding boxes and labels
• 3-classes annotation is available: with_mask, without_mask, mask_weared_incorrect (not used in this project)

Folder Structure

FaceMaskDetection

|-- data
|    |-- original_data |    |    |-- images
|    |    |-- annotations |    |-- train
|    |    |-- images
|    |    |-- annotations |    |-- test |    |    |-- images
|    |    |-- annotations
|-- utilities
|-- output
|-- model
README.md
requirements.txt
cleanup.py
train.py
test.py

Env creation (one time)

virtualenv venv3/bin/activate

source venv3/bin/activate

pip3 install -r requirements.txt

pip3 install jupyter notebook

pip install ipykernel

python -m ipykernel install --user --name=venv3

deactivate

source venv3/bin/activate

Activating env (every time u open terminal)

source venv3/bin/activate

jupyter notebook

FRCNN

FRCNN_colab.ipynb

Classifier training

cnn_classifier_train.ipynb

Change experiemnt name in colab (see top cell)

• the testing results will be written into the output folder, here's an example of prediction: