It is a simple demo including face detection and face aligment, and some optimizations were made to make the result better.
click the gif to see the video:
- PyTorch
- onnxruntime
- opencv
Simple keypoints detector.
WFLW | NME | Flops(G) | Params(M) | Pose | Exp. | Ill. | Mu. | Occ. | Blur | pretrained |
---|---|---|---|---|---|---|---|---|---|---|
Student@128 | 4.80 | 0.35 | 3.25 | 8.53 | 5.00 | 4.61 | 4.81 | 5.80 | 5.36 | skps |
Teacher@128 | 4.17 | 1.38 | 11.53 | 7.14 | 4.32 | 4.01 | 4.03 | 4.98 | 4.68 | skps |
Student@256 | 4.35 | 1.39 | 3.25 | 7.53 | 4.52 | 4.16 | 4.21 | 5.34 | 4.93 | skps |
Teacher@256 | 3.95 | 5.53 | 11.53 | 7.00 | 4.00 | 3.81 | 3.78 | 4.85 | 4.54 | skps |
I will release new model when there is better one.
By default student@256 is used in this project
git clone https://github.com/610265158/Peppa_Pig_Face_Landmark
python setup.py install
# by code:
from Skps import FaceAna
facer = FaceAna()
boxes, landmarks, _ = facer.run(image)
## detect images, tracing is not used, add
## facer.reset()
More details refer to demo.py
run python demo.py --cam_id 0
use a camera
or python demo.py --video test.mp4
detect for a video
or python demo.py --img_dir ./test
detect for images dir no track
or python demo.py --video test.mp4 --mask True
if u want a face mask
The codes are placed in folder TRAIN/face_landmark
Refer to TRAIN/face_landmark/README.md to train the model.