Super intéressant et de meilleur qualité que ce que j'avais testé il y a 2ans.
MegaPortraits: One-shot Megapixel Neural Head Avatars
It samples two random frames from the dataset at each step: the source frame and the driver frame. The model imposes the motion of the driving frame (i.e., the head pose and the facial expression) onto the appearance of the source frame to produce an output image.
The main learning signal is obtained from the training episodes where the source and the driver frames come from the same video, and hence our model’s prediction is trained to match the driver frame.
via LinkedIn : lire l’article source