Face-Object-Detection

Face detection model train with model YOLOv8

Trained model

The model was trained using a NVIDIA GeForce GTX 1070. An example using the trained model on images from the validation set is shown below.

 

Dataset and preprocessing tools

FDDB (Face Detection Data Set) was used to train the model. The script tools/preprocess_dataset.py process and transform images and labels to the yolov8 dataset format.

./tools/preprocess_dataset.py --dataset_root_path --save_root_path

Training metrics

Training metrics are shown bellow with P, R, F1 and PR curves.