Knowledge-aware Bayesian Co-attention for Multimodal Emotion Recognition
This work addresses the challenge of learning emotionally relevant parts in multimodal emotion recognition, which is important for applications in human-computer interaction and affective computing, though it is incremental as it builds on existing attention-based models.
The paper tackled the problem of multimodal emotion recognition by incorporating external emotion-related knowledge into a co-attention model, resulting in an improvement of at least 0.7% unweighted accuracy over state-of-the-art methods on the IEMOCAP dataset.
Multimodal emotion recognition is a challenging research area that aims to fuse different modalities to predict human emotion. However, most existing models that are based on attention mechanisms have difficulty in learning emotionally relevant parts on their own. To solve this problem, we propose to incorporate external emotion-related knowledge in the co-attention based fusion of pre-trained models. To effectively incorporate this knowledge, we enhance the co-attention model with a Bayesian attention module (BAM) where a prior distribution is estimated using the emotion-related knowledge. Experimental results on the IEMOCAP dataset show that the proposed approach can outperform several state-of-the-art approaches by at least 0.7% unweighted accuracy (UA).