I am an experienced Deep Learning Engineer, currently working as an AI engineer at TU Munich, Germany.
From what I understand, you want two models to be trained: Facial pose estimation and a GAN for super-resolution.
I can provide some insight to you for these tasks before hand. This paper : ([login to view URL]) is one of the beast for facial pose estimation to my knowledge. I have read this paper for my research and is implementable. For super-resolution, many GAN models have been proposed and I can look deeply into this as soon as I have been awarded the project.
I have done research in the field of Computer Vision at IIT Delhi for 2 years, so I am well aware of the work that needs to go in this task. Needless to say, you don't have to worry for the outcome, my work is full-proof and your expectations will be met!
Looking forward to hearing from you.