Beyond Appearance: Transformer-based Person Identification from Conversational Dynamics
This work addresses person identification in natural interactions, offering incremental improvements for applications in social computing and multimodal analysis.
The paper tackled person identification in face-to-face conversations by using a transformer-based two-stream framework to model spatial and temporal keypoint data, achieving 98.03% accuracy with feature-level fusion.
This paper investigates the performance of transformer-based architectures for person identification in natural, face-to-face conversation scenario. We implement and evaluate a two-stream framework that separately models spatial configurations and temporal motion patterns of 133 COCO WholeBody keypoints, extracted from a subset of the CANDOR conversational corpus. Our experiments compare pre-trained and from-scratch training, investigate the use of velocity features, and introduce a multi-scale temporal transformer for hierarchical motion modeling. Results demonstrate that domain-specific training significantly outperforms transfer learning, and that spatial configurations carry more discriminative information than temporal dynamics. The spatial transformer achieves 95.74% accuracy, while the multi-scale temporal transformer achieves 93.90%. Feature-level fusion pushes performance to 98.03%, confirming that postural and dynamic information are complementary. These findings highlight the potential of transformer architectures for person identification in natural interactions and provide insights for future multimodal and cross-cultural studies.