Then a comparison between these two returned lists is done by the function compare_faces() which returns a list of boolean values(True or False). Both these two steps are followed for the original and test image. Then face encodings(markings of eyes, nose, mouth, jaws which remain the same for different images of the same person) are taken using face_encodings() function which returns a list containing 128 measurements.

The last step is to match these encoding with the nearest possible image from a stored database.įirst, we get the location of where exactly the face is in the image using face_location() method(which gets the outline of the face) on the RGB image. Source – face recognition library documentation Confusing between similar looking peopleįace recognition algorithms can extract features from a face image namely positions of forehead, eyes, nose, mouth, chin, jaws.įace Landmarks – There are 68 specific points (called landmarks) that exist on every face.įace Encodings – This is the 128 encoding feature vector from a pretrained network over millions of images.Differently posing – there could be images of the same person with different face angles.