CoMPM: Context Modeling with Speaker's Pre-trained Memory Tracking for Emotion Recognition in Conversation.
Joosung LeeWooin LeePublished in: NAACL-HLT (2022)
Keyphrases
- emotion recognition
- context modeling
- audio visual
- pre trained
- multi modal
- training data
- particle filter
- human computer interaction
- facial expressions
- visual information
- visual data
- training examples
- object tracking
- multimedia
- natural language
- multiscale
- artificial intelligence
- motion estimation
- appearance model
- facial images
- object detection
- three dimensional
- high resolution