Methods and systems for realistic head rotations and facial animation compositing on a mobile device

摘要

Methods and systems are provided for realistic head rotations and facial animation synthesis. An example method may include receiving frames of a source video having a face and head of a source actor. The method may then proceed to generating sets of source pose parameters representing the facial expressions of the source actor and positions of the head. The method includes receiving at least one target image comprising a target face and a target head of a target person, determining target identity information associated with the target face, and based on sets of source pose parameters and target identity information. The method may further include generating an output video. Each frame of the output video may include an image of the target face modified to mimic at least one of the source actor''s facial expressions and at least one of the source actor''s head positions.