Social-MAE: A Transformer-Based Multimodal Autoencoder for Face and Voice.
Hugo BohyMinh TranKevin El HaddadThierry DutoitMohammad SoleymaniPublished in: FG (2024)
Keyphrases
- voice recognition
- social interaction
- human faces
- face biometrics
- multi modal
- social networks
- fuzzy logic
- multimodal biometrics
- face images
- social media
- social context
- power system
- social learning
- online social networks
- distribution network
- multimodal interaction
- multimodal fusion
- incipient fault
- social relationships
- social networking sites
- online communities
- social networking
- lighting conditions
- principal component analysis
- facial expressions
- semi supervised
- expert systems
- multimedia