pytorch implementation of face detection algorithm MTCNN
Just download the repository and then do this
from src.detector import detect_faces
from src.utils import show_bboxes
from PIL import Image
image = Image.open('images/test3.jpg')
bounding_boxes, landmarks = detect_faces(image)
image = show_bboxes(image, bounding_boxes, landmarks)
image.show()
- pytorch 0.4
- Pillow, numpy
This implementation is heavily inspired by:
- pangyupo/mxnet_mtcnn_face_detection
- https://github.com/kpzhang93/MTCNN_face_detection_alignment
- https://github.com/TropComplique/mtcnn-pytorch
MTCNN: Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks.