- Make a pip env
python3 -m venv ./venv
- Activate the env
source ./venv/bin/activate
- Install ultralytics
pip install ultralytics
- Use infer.py module for inference
python infer.py
- I made a dataset of over 20k images containing people.
- I get ground-truth bounding boxes using MTCNN
- Trained YOLOv8 on the dataset for face detection
Please note that the dataset and the training procedure should be improved for boosting the performance.
A big 👍 for ultralytics