87 lines
2.6 KiB
Python
87 lines
2.6 KiB
Python
import os
|
|
import bz2
|
|
import gdown
|
|
import numpy as np
|
|
from deepface.commons import functions
|
|
from deepface.commons.logger import Logger
|
|
|
|
logger = Logger(module="basemodels.DlibResNet")
|
|
|
|
# pylint: disable=too-few-public-methods
|
|
|
|
|
|
class DlibResNet:
|
|
def __init__(self):
|
|
|
|
## this is not a must dependency. do not import it in the global level.
|
|
try:
|
|
import dlib
|
|
except ModuleNotFoundError as e:
|
|
raise ImportError(
|
|
"Dlib is an optional dependency, ensure the library is installed."
|
|
"Please install using 'pip install dlib' "
|
|
) from e
|
|
|
|
self.layers = [DlibMetaData()]
|
|
|
|
# ---------------------
|
|
|
|
home = functions.get_deepface_home()
|
|
weight_file = home + "/.deepface/weights/dlib_face_recognition_resnet_model_v1.dat"
|
|
|
|
# ---------------------
|
|
|
|
# download pre-trained model if it does not exist
|
|
if os.path.isfile(weight_file) != True:
|
|
logger.info("dlib_face_recognition_resnet_model_v1.dat is going to be downloaded")
|
|
|
|
file_name = "dlib_face_recognition_resnet_model_v1.dat.bz2"
|
|
url = f"http://dlib.net/files/{file_name}"
|
|
output = f"{home}/.deepface/weights/{file_name}"
|
|
gdown.download(url, output, quiet=False)
|
|
|
|
zipfile = bz2.BZ2File(output)
|
|
data = zipfile.read()
|
|
newfilepath = output[:-4] # discard .bz2 extension
|
|
with open(newfilepath, "wb") as f:
|
|
f.write(data)
|
|
|
|
# ---------------------
|
|
|
|
model = dlib.face_recognition_model_v1(weight_file)
|
|
self.__model = model
|
|
|
|
# ---------------------
|
|
|
|
# return None # classes must return None
|
|
|
|
def predict(self, img_aligned: np.ndarray) -> np.ndarray:
|
|
|
|
# functions.detectFace returns 4 dimensional images
|
|
if len(img_aligned.shape) == 4:
|
|
img_aligned = img_aligned[0]
|
|
|
|
# functions.detectFace returns bgr images
|
|
img_aligned = img_aligned[:, :, ::-1] # bgr to rgb
|
|
|
|
# deepface.detectFace returns an array in scale of [0, 1]
|
|
# but dlib expects in scale of [0, 255]
|
|
if img_aligned.max() <= 1:
|
|
img_aligned = img_aligned * 255
|
|
|
|
img_aligned = img_aligned.astype(np.uint8)
|
|
|
|
model = self.__model
|
|
|
|
img_representation = model.compute_face_descriptor(img_aligned)
|
|
|
|
img_representation = np.array(img_representation)
|
|
img_representation = np.expand_dims(img_representation, axis=0)
|
|
|
|
return img_representation
|
|
|
|
|
|
class DlibMetaData:
|
|
def __init__(self):
|
|
self.input_shape = [[1, 150, 150, 3]]
|