Dyadic Interaction
Dyadic interaction research focuses on understanding the complex interplay between two individuals, aiming to model and predict their behaviors and interactions across various modalities (audio, video, text). Current research emphasizes developing sophisticated models, including neural networks (e.g., transformers, neural operators) and graph-based approaches, to analyze these interactions for applications such as mental health diagnosis, human-robot interaction, and personalized recommendations. These advancements have significant implications for fields like psychology, computer science, and healthcare, enabling more accurate assessments of social behavior, improved human-computer interaction design, and more effective therapeutic interventions.
Papers
Loose Social-Interaction Recognition in Real-world Therapy Scenarios
Abid Ali, Rui Dai, Ashish Marisetty, Guillaume Astruc, Monique Thonnat, Jean-Marc Odobez, Susanne Thümmler, Francois Bremond
Active Listener: Continuous Generation of Listener's Head Motion Response in Dyadic Interactions
Bishal Ghosh, Emma Li, Tanaya Guha
Multimodal Dyadic Impression Recognition via Listener Adaptive Cross-Domain Fusion
Yuanchao Li, Peter Bell, Catherine Lai
Perceived personality state estimation in dyadic and small group interaction with deep learning methods
Kristian Fenech, Ádám Fodor, Sean P. Bergeron, Rachid R. Saboundji, Catharine Oertel, András Lőrincz