CLIP-PAE: Projection-Augmentation Embedding to Extract Relevant Features for a Disentangled, Interpretable and Controllable Text-Guided Face Manipulation.
Chenliang ZhouFangcheng ZhongCengiz ÖztireliPublished in: SIGGRAPH (Conference Paper Track) (2023)
Keyphrases
- automatically extracted
- feature extraction
- video clips
- classification accuracy
- information retrieval
- co occurrence
- content features
- feature set
- person authentication
- text retrieval
- lexical features
- facial expressions
- low level
- highly informative
- feature space
- keypoints
- additional features
- video segments
- vector space
- wordnet
- feature vectors
- search engine