«YOLO8Face» combined YOLO5Face and YOLOv8 for face and keypoint detection

Through experiments, it was found that using YOLOv8-Pose can simultaneously detect faces and facial keypoints. Using the facial and keypoint datasets provided by Retina Face, only the training dataset was used. Validation/training was split into 2:8 parts, and the facial and keypoint detectors were trained from scratch. Finally, they were evaluated on the WIDERFACE val dataset.

	ARCH	GFLOPs	Easy	Medium	Hard
zjykzj/YOLO5Face	yolov5s-v7.0	15.8	94.84	93.28	84.67
zjykzj/YOLO5Face	yolov5n-v7.0	4.2	93.25	91.11	80.33
zjykzj/YOLO8Face	yolov5su	23.8	95.18	93.50	82.47
zjykzj/YOLO8Face	yolov5nu	7.1	93.96	91.82	78.89
zjykzj/YOLO8Face	yolov8s	28.4	95.81	94.26	82.75
zjykzj/YOLO8Face	yolov8n	8.1	94.57	92.55	78.97
zjykzj/YOLO8Face	yolov8s-pose	29.4	95.07	93.77	82.84
zjykzj/YOLO8Face	yolov8n-pose	8.3	94.07	92.04	78.88

During the eval phase, using VGA resolution input images (the longer edge of the input image is scaled to 640, and the shorter edge is scaled accordingly)

Table of Contents

• Table of Contents
• News🚀🚀🚀
• Background
• Installation
• Usage
  • Train
  • Eval
  • Predict
• Maintainers
• Thanks
• Contributing
• License

News🚀🚀🚀

Version	Release Date	Major Updates
v0.1.0	2025/02/03	Trains a face detector based on YOLOv8 and the WIDERFACE dataset.

Background

According to the YOLO5Face implementation, it adds Landmarks-HEAD to YOLOv5 to achieve synchronous detection of faces and keypoints. The YOLOv8 is an upgraded version of YOLOv5, which naturally improves the performance of face and keypoint detection by combining YOLO5Face and YOLOv8.

Note: the latest implementation of YOLO8Face in our warehouse is entirely based on ultralytics/ultralytics v8.2.103

Installation

See INSTALL.md

Usage

Download the WIDERFACE dataset from http://shuoyang1213.me/WIDERFACE/, Then convert WIDERFACE dataset format.

$ python3 widerface2yolo-pose.py ../datasets/widerface/WIDER_train/images ../datasets/widerface/retinaface_gt_v1.1/train/label.txt ../datasets/widerface-landmarks/

Train

# YOLOv8n-pose / YOLOv8s-pose
$ python3 pose_train.py --model yolov8n-pose.pt --data ./yolo8face/cfg/datasets/widerface-landmarks.yaml --epochs 100 --imgsz 640 --device 0
$ python3 pose_train.py --model yolov8s-pose.pt --data ./yolo8face/cfg/datasets/widerface-landmarks.yaml --epochs 100 --imgsz 640 --device 0

Eval

# python pose_widerface.py --model yolov8s-pose_widerface.pt --source ../datasets/widerface/images/val/ --folder_pict ../datasets/widerface/wider_face_split/wider_face_val_bbx_gt.txt --save_txt true --conf 0.001 --iou 0.7 --max_det 300 --batch 1 --device 0
args: Namespace(data=None, device=[0], folder_pict='../datasets/widerface/wider_face_split/wider_face_val_bbx_gt.txt', model='yolov8s-pose_widerface.pt', source='../datasets/widerface/images/val/') - unknown: ['--save_txt', 'true', '--conf', '0.001', '--iou', '0.7', '--max_det', '300', '--batch', '1']
{'model': 'yolov8s-pose_widerface.pt', 'data': None, 'device': [0], 'source': '../datasets/widerface/images/val/', 'folder_pict': '../datasets/widerface/wider_face_split/wider_face_val_bbx_gt.txt', 'save_txt': True, 'conf': 0.001, 'iou': 0.7, 'max_det': 300, 'batch': 1, 'mode': 'predict'}
3226

Ultralytics YOLOv8.2.103 🚀 Python-3.8.19 torch-1.12.1+cu113 CUDA:0 (NVIDIA GeForce RTX 3090, 24268MiB)
YOLOv8s-pose summary (fused): 187 layers, 11,413,344 parameters, 0 gradients, 29.4 GFLOPs
...
...
Speed: 2.1ms preprocess, 12.9ms inference, 1.6ms postprocess per image at shape (1, 3, 640, 448)
Results saved to /data/zj/YOLO8Face/runs/detect/predict
0 label saved to /data/zj/YOLO8Face/runs/detect/predict/labels
# cd widerface_evaluate/
# python3 evaluation.py -p ../runs/detect/predict/labels/ -g ./ground_truth/
Reading Predictions : 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 61/61 [00:00<00:00, 116.17it/s]
Processing easy: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 61/61 [00:18<00:00,  3.24it/s]
Processing medium: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 61/61 [00:18<00:00,  3.24it/s]
Processing hard: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 61/61 [00:19<00:00,  3.21it/s]
==================== Results ====================
Easy   Val AP: 0.9507304597200303
Medium Val AP: 0.9376731961249901
Hard   Val AP: 0.8283959392916
=================================================

Predict

# python3 pose_predict.py --model yolov8s-pose_widerface.pt --source ./yolo8face/assets/widerface_val/ --imgsz 640 --device 0
args: Namespace(data=None, device=[0], model='yolov8s-pose_widerface.pt', source='./yolo8face/assets/widerface_val/') - unknown: ['--imgsz', '640']

Ultralytics YOLOv8.2.103 🚀 Python-3.8.19 torch-1.12.1+cu113 CUDA:0 (NVIDIA GeForce RTX 3090, 24268MiB)
YOLOv8s-pose summary (fused): 187 layers, 11,413,344 parameters, 0 gradients, 29.4 GFLOPs
image 1/2 /data/zj/YOLO8Face/yolo8face/assets/widerface_val/39_Ice_Skating_iceskiing_39_351.jpg: 640x640 3 faces, 10.0ms
image 2/2 /data/zj/YOLO8Face/yolo8face/assets/widerface_val/9_Press_Conference_Press_Conference_9_632.jpg: 640x640 1 face, 10.0ms
Speed: 3.7ms preprocess, 10.0ms inference, 1.5ms postprocess per image at shape (2, 3, 640, 640)
Results saved to /data/zj/YOLO8Face/runs/detect/predict4

Maintainers

• zhujian - Initial work - zjykzj

Thanks

• ultralytics/ultralytics
• zjykzj/YOLO5Face
• deepcam-cn/yolov5-face

Contributing

Anyone's participation is welcome! Open an issue or submit PRs.

License

Apache License 2.0 © 2025 zjykzj