fdet
fdet copied to clipboard
An easy to use face detection module based on MTCNN and RetinaFace algorithms.
testing below: import cv2 from fdet import io, RetinaFace BATCH_SIZE = 1000 detector = RetinaFace(backbone='MOBILENET', cuda_devices=[0,1,2,3]) vid_cap = cv2.VideoCapture('13.mp4') video_face_detections = [] # list to store all video face detections...
RESNET50 and MOBILENET model files at drop box are not found
``` def _load_mobile_net_model(self): model = _MobileNetV1() state_dict = load_state_dict_from_url( 'https://www.dropbox.com/s/bd1keyo085pscfu/mobilenetv1_pretrain.pt?dl=1', map_location=self._device_control ) # # load params model.load_state_dict(state_dict) return model`` ``` this dropbox link for mobile net is not working