Abstract: An animation generation method for tracking a facial expression and a neural network training method thereof are provided. The animation generation method for tracking a facial expression includes: driving a first role model according to an expression parameter set to obtain a virtual expression image corresponding to the expression parameter set; applying a plurality of real facial images to the virtual expression image corresponding to the facial expression respectively to generate a plurality of real expression images; training a tracking neural network according to the expression parameter set and the real expression images; inputting a target facial image to the trained tracking neural network to obtain a predicted expression parameter set; and using the predicted expression parameter set to control a second role model.
Type:
Grant
Filed:
March 8, 2022
Date of Patent:
April 9, 2024
Assignee:
DIGITAL DOMAIN ENTERPRISES GROUP LIMITED