Exploiting modality-invariant feature for robust multimodal emotion recognition with missing modalities.
Haolin ZuoRui LiuJinming ZhaoGuanglai GaoHaizhou LiPublished in: CoRR (2022)
Keyphrases
- emotion recognition
- audio visual
- multi modal
- invariant features
- multiple modalities
- single modality
- emotional speech
- facial expressions
- human computer interaction
- sentiment analysis
- multi stream
- visual data
- image processing
- emotional state
- text mining
- image registration
- high dimensional
- image sequences
- emotion classification
- multimedia