Leveraging Recent Advances in Deep Learning for Audio-Visual Emotion Recognition
This work addresses emotion recognition for human behavior analysis, offering incremental improvements in performance for specific datasets.
The paper tackles audio-visual emotion recognition by proposing a deep learning approach that fuses audio and visual features with a model-level strategy and uses a recurrent neural network for temporal dynamics, achieving state-of-the-art results on the RECOLA dataset for valence prediction and outperforming benchmarks on AffectNet and Google Facial Expression Comparison datasets.
Emotional expressions are the behaviors that communicate our emotional state or attitude to others. They are expressed through verbal and non-verbal communication. Complex human behavior can be understood by studying physical features from multiple modalities; mainly facial, vocal and physical gestures. Recently, spontaneous multi-modal emotion recognition has been extensively studied for human behavior analysis. In this paper, we propose a new deep learning-based approach for audio-visual emotion recognition. Our approach leverages recent advances in deep learning like knowledge distillation and high-performing deep architectures. The deep feature representations of the audio and visual modalities are fused based on a model-level fusion strategy. A recurrent neural network is then used to capture the temporal dynamics. Our proposed approach substantially outperforms state-of-the-art approaches in predicting valence on the RECOLA dataset. Moreover, our proposed visual facial expression feature extraction network outperforms state-of-the-art results on the AffectNet and Google Facial Expression Comparison datasets.