Invisible Strings: Revealing Latent Dancer-to-Dancer Interactions with Graph Neural Networks

2025-03-04

AI Summary

This study addresses the challenge of modeling and interpreting implicit, non-contact interpersonal interactions in partner dance. Methodologically, high-fidelity 3D pose sequences are extracted from dance videos to construct temporal dynamic relational graphs; for the first time, graph neural networks (GNNs) are employed to model invisible biomechanical coupling and perceptual coordination between dancers. A novel interpretable interaction weight prediction framework is proposed, integrating motion reconstruction optimization with interactive visualization. Key contributions include: (1) identification of canonical non-contact interaction patterns—such as mirroring guidance, momentum transfer, and spatial tension; (2) achieving high predictive consistency on real-world contemporary dance excerpts; and (3) bridging embodied artistic practice with computational modeling, thereby enabling the development of generative tools for dance rehearsal and choreographic assistance.

Technology Category

• Humans and AI: Interaction Techniques and Devices
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Computer Vision: Video Understanding & Activity Analysis

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Dancing in a duet often requires a heightened attunement to one's partner: their orientation in space, their momentum, and the forces they exert on you. Dance artists who work in partnered settings might have a strong embodied understanding in the moment of how their movements relate to their partner's, but typical documentation of dance fails to capture these varied and subtle relationships. Working closely with dance artists interested in deepening their understanding of partnering, we leverage Graph Neural Networks (GNNs) to highlight and interpret the intricate connections shared by two dancers. Using a video-to-3D-pose extraction pipeline, we extract 3D movements from curated videos of contemporary dance duets, apply a dedicated pre-processing to improve the reconstruction, and train a GNN to predict weighted connections between the dancers. By visualizing and interpreting the predicted relationships between the two movers, we demonstrate the potential for graph-based methods to construct alternate models of the collaborative dynamics of duets. Finally, we offer some example strategies for how to use these insights to inform a generative and co-creative studio practice.

Problem

• Capturing subtle dancer interactions in duets
• Using GNNs to model dancer relationships
• Enhancing dance documentation and studio practice

Innovation

• Graph Neural Networks analyze dancer interactions
• Video-to-3D-pose extraction captures dance movements
• Weighted connections predict collaborative dynamics