Multimodal Multi-View Spectral-Spatial-Temporal Masked Autoencoder for Self-Supervised Emotion Recognition.
Pengxuan GaoTianyu LiuJia-Wen LiuBao-Liang LuWei-Long ZhengPublished in: ICASSP (2024)
Keyphrases
- multi view
- spatial temporal
- emotion recognition
- audio visual
- multi modal
- action recognition
- spatio temporal
- human computer interaction
- spatial and temporal
- d objects
- three dimensional
- facial expressions
- video shots
- visual information
- semi supervised
- emotional state
- visual data
- multimedia
- context aware
- image data
- viewpoint
- facial images
- image sequences