I have a video processing script but the output videos generated has some issues around the lips area (see attachment). The script is written in Python and hosted on AWS cloud.
I'm looking for someone who is expert in the video processing field to provide an automated solution that will enhance the facial features and blurry area around the mouth for all the generated output videos. The solution needs to be lightweight and integrated in our workflow and should return a one minute of minute back to us fast (if not real-time).
Perhaps with maybe some face parsing + super resolution or cv2 interpolation?