MAGIC-TBR: Multiview Attention Fusion for Transformer-based Bodily Behavior Recognition in Group Settings

TL;DR

This paper introduces MAGIC-TBR, a transformer-based multiview attention fusion method that improves the detection of fine bodily behaviors in group settings by combining video features and DCT coefficients.

Contribution

It presents a novel multiview attention fusion approach for bodily behavior recognition, integrating video and DCT features within a transformer framework.

Findings

Effective feature fusion demonstrated on BBSI dataset

Improved recognition of fine bodily behaviors

Transformer-based approach outperforms existing methods

Abstract

Bodily behavioral language is an important social cue, and its automated analysis helps in enhancing the understanding of artificial intelligence systems. Furthermore, behavioral language cues are essential for active engagement in social agent-based user interactions. Despite the progress made in computer vision for tasks like head and body pose estimation, there is still a need to explore the detection of finer behaviors such as gesturing, grooming, or fumbling. This paper proposes a multiview attention fusion method named MAGIC-TBR that combines features extracted from videos and their corresponding Discrete Cosine Transform coefficients via a transformer-based approach. The experiments are conducted on the BBSI dataset and the results demonstrate the effectiveness of the proposed feature fusion with multiview attention. The code is available at: https://github.com/surbhimadan92/MAGIC-TBR