I am using Keras to generate pictures to feed face_recognition package.
the following code I used to read and prepare the picture to be passed to the generator
image = cv2.imread(imagePath)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image = np.expand_dims(image, axis=0)
imageGen = aug.flow(image, batch_size=1)
then for the generated image is used as follow:
for gimage in imageGen:
face_recognition.face_locations(gimage, 'cnn')
RuntimeError: Unsupported image type, must be 8bit gray or RGB image.
I tried to solve the issue by using squeeze
before passing the generated picture; but also it did not work
gimage = np.squeeze(gimage, axis=0)
You should check gimage ndarray's value dtype. It should be uint8
and as per the error it looks like currently it is float32