We worked on detecting and recognising the face features using custom collected dataset. We have used PyTorch as deep learning Framework and have used ResNet weights to finetune the model with five classes in our dataset.
Live inference can be found at huggingface
We have used python 3.11 Version to run this game on and latest pytorch environment
First run this command in a virtual environment to install the dependencies
pip install -r requirements.txtAfter setting up the environment, Run
python train.pyThis will start training the model and save it to outputs directory.
Once the model is trained, you can do the inference or deploy it to the gradio app. I have written code for deploying it through gradio and inference the model with input. After training the model run
python webapp.py