Social-MAE: A Transformer-Based Multimodal Autoencoder for Face and Voice
This work addresses the need for powerful audiovisual models in social and affective computing, though it is incremental as it builds on an existing method with modifications.
The authors tackled the problem of multimodal perception of human social behaviors by developing Social-MAE, a transformer-based masked autoencoder pre-trained on audiovisual social data, which achieved state-of-the-art results on emotion and laughter recognition and competitive results on personality estimation.
Human social behaviors are inherently multimodal necessitating the development of powerful audiovisual models for their perception. In this paper, we present Social-MAE, our pre-trained audiovisual Masked Autoencoder based on an extended version of Contrastive Audio-Visual Masked Auto-Encoder (CAV-MAE), which is pre-trained on audiovisual social data. Specifically, we modify CAV-MAE to receive a larger number of frames as input and pre-train it on a large dataset of human social interaction (VoxCeleb2) in a self-supervised manner. We demonstrate the effectiveness of this model by finetuning and evaluating the model on different social and affective downstream tasks, namely, emotion recognition, laughter detection and apparent personality estimation. The model achieves state-of-the-art results on multimodal emotion recognition and laughter recognition and competitive results for apparent personality estimation, demonstrating the effectiveness of in-domain self-supervised pre-training. Code and model weight are available here https://github.com/HuBohy/SocialMAE.