yolo5face 0.0.4

Wrapper over Yolo5Face for a more convenient inference.

Wrapper over YoloV5Face

Installation

pip install -U yolo5face

Inference

from yolo5face.get_model import get_model

model = get_model("yolov5n", gpu=-1, target_size=512, min_face=24)

image = cv2.imread(<IMAGE_PATH>)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

boxes, key_points = model(image)

• gpu - GPU number, -1 or cpu for CPU
• target_size - min size of the target_image
• min_face - minimum face size in pixels. All faces that have side smaller than min_face will be ignored.