Fork of derronqi/yolov7-face

Differences between original repository and fork:

• Compatibility with PyTorch >=2.5. (🔥)
• Original pretrained models and converted ONNX models from GitHub releases page. (🔥)
• The wider_val.txt file for WIDERFace evaluation.
• The following deprecations 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.

Installation

shell pip install -r requirements.txt

Pretrained models

• Download links:

| Name | Model Size (MB) | Link | SHA-256 | |--------------------|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------| | YOLOv7-Lite-t-Face | 0.7
2.1 | PyTorch
ONNX | 04ea1dedc42c336004a29a770da8cbc795e50b0d835366b611578fd7648a1f3a
448b3ad27119ca15f629d9e83fe4e8c71a783c0358a154f5f4f9197320a0c89d | | YOLOv7-Lite-s-Face | 2.3
5.2 | PyTorch
ONNX | 8ad12d062bf14ec27196aa159312c33347f5850cfc4bf6e977725404ac64f01d
0570fb8472d2eba1350612c0d4d691abe1337e3f173069438940dc3686f0e335 | | YOLOv7-Tiny-Face | 11.8
24.2 | PyTorch
ONNX | 91eab824696f6cfadbc60bee63e969d4b4ca0484bc4c1e57b4dbd73917990127
76c213da5cda623984ab8c0ca97cf30fec1f75ced7fff9a0e1a1afbda8efdbc0 | | YOLOv7s-Face | 8.5
17.4 | PyTorch
ONNX | 044a2194cbb646ff8c31cc4264a10e061dc269929ac7b881db292511fc43c2c9
e69b61c08ab856766c65605aabc7ed6620fcdc9ab9b5d1fb2c558ed429c8ff3b | | YOLOv7-Face | 70.2
140.6 | PyTorch
ONNX | 20dd9c9834b467be1b1302b089c3b97d2d0055bae99c97f69d1a5bb63323911b
2c588b1bad756c189257d6de813837b04931441c10dae789121e5e1f95da1f64 | | YOLOv7-W6-Face | 133.9
267.8 | PyTorch
ONNX | 61b6205a6e01632b0df2bb6775420d65657676fa528b817966fbb6519afdca4a
6263fd8022268a029573300e1de273db3b091b5b745520e88e735b2faa296c9a |

• Evaluation results on WIDERFace dataset:

| Name | Image Size | Easy | Medium | Hard | FLOPs (B) @640 | |--------------------|------------|------|--------|------|----------------| | YOLOv7-Lite-t-Face | 640 | 88.7 | 85.2 | 71.5 | 0.8 | | YOLOv7-Lite-s-Face | 640 | 92.7 | 89.9 | 78.5 | 3.0 | | YOLOv7-Tiny-Face | 640 | 94.7 | 92.6 | 82.1 | 13.2 | | YOLOv7s-Face | 640 | 94.8 | 93.1 | 85.2 | 16.8 | | YOLOv7-Face | 640 | 96.9 | 95.5 | 88.0 | 103.4 | | YOLOv7-W6-Face | 960 | 96.4 | 95.0 | 88.3 | 89.0 |

Inference

shell python detect.py --weights yolov7s-face.pt --source data/images/22_Picnic_Picnic_22_10.jpg

WIDERFace evaluation

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

shell python test_widerface.py --weights yolov7s-face.pt --dataset_folder data/widerface/val/images/ shell cd widerface_evaluate shell python setup.py build_ext --inplace shell python evaluation.py

Export to ONNX format

shell pip install onnx shell python models/export.py --weights yolov7s-face.pt --grid