VAST: Vivify Your Talking Avatar via Zero-Shot Expressive Facial Style Transfer.
Liyang ChenZhiyong WuRunnan LiWeihong BaoJun LingXu TanSheng ZhaoPublished in: CoRR (2023)
Keyphrases
- facial animation
- facial images
- facial expressions
- facial motion
- face recognition
- virtual world
- transfer learning
- human computer interaction
- emotion recognition
- knowledge transfer
- behavior recognition
- landmark detection
- human faces
- database
- cross domain
- facial expression recognition
- facial recognition
- decision trees
- video sequences
- feature extraction
- authorship attribution
- virtual reality
- active learning
- multiscale
- website
- computer vision
- machine learning