Fork of Krasjet-Yu/YOLO-FaceV2

Differences between original repository and fork:

• Compatibility with PyTorch >=2.4. (🔥)
• Converted ONNX models from GitHub releases page. (🔥)
• The wider_val.txt file for WIDERFace evaluation.
• The following deprecations and errors has been fixed:
  • UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument.
  • DeprecationWarning: 'np.float' is a deprecated alias for builtin 'float'.
  • FutureWarning: You are using 'torch.load' with 'weights_only=False'.
  • FutureWarning: Cython directive 'language_level' not set.
  • Cython Warning: Using deprecated NumPy API.
  • AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'.

Installation

pip install -r requirements.txt

Pretrained models

• Download links:

Name	Model Size (MB)	Link	SHA-256
YOLO-FaceV2	36.2 71.8	PyTorch ONNX	f424fd437b22711207c48aac4c205d844eaea2f3c12e2f3f7ecd7f9650634e44 8ad769e4e9fd5869baa4243367e041df901a228e121a1602deec85074d2bd7ed

Inference

python detect.py --weights yolo-facev2.pt --source data/images/bus.jpg

WIDERFace evaluation

• Download WIDERFace validation dataset.
• Move dataset to data/widerface/val directory.

python widerface_pred.py --weights yolo-facev2.pt --dataset_folder data/widerface/val/images

cd widerface_evaluate

python setup.py build_ext --inplace

python evaluation.py

Export to ONNX format

pip install onnx

python models/export.py --weights yolo-facev2.pt --grid