Yolov11n Face Detection

🚀 YOLOv11n-Face-Detection

A lightweight face detection model based on the YOLO architecture, offering high - efficiency face detection capabilities.

🚀 Quick Start

This is a lightweight face detection model built upon the YOLO architecture, specifically YOLOv11 nano. It was trained for 225 epochs on the WIDERFACE dataset.

The model achieves the following results on the evaluation set:

==================== Results ====================
Easy   Val AP: 0.9420471677096086
Medium Val AP: 0.9210357271019756
Hard   Val AP: 0.8099848364072022
=================================================

Here are the YOLO results:

The Confusion matrix is as follows:

[[23577 2878]

[16098 0]]

💻 Usage Examples

Basic Usage

from huggingface_hub import hf_hub_download
from ultralytics import YOLO

model_path = hf_hub_download(repo_id="AdamCodd/YOLOv11n-face-detection", filename="model.pt")
model = YOLO(model_path)

results = model.predict("/path/to/your/image", save=True) # saves the result in runs/detect/predict

🔧 Technical Details

Limitations

• The model's performance may vary under extreme lighting conditions.
• It is best suited for frontal and slightly angled faces.
• Optimal performance is achieved when faces occupy more than 20 pixels.

📄 License

This project is licensed under the Apache - 2.0 license.