UDIVA HHOI Challenge and CONTEXTUS at ECCV 2026. Co-organised the challenge and workshop on understanding human interaction in context.

New SERMAS paper in the International Journal of Human-Computer Studies. The paper presents a modular architecture for context-aware, socially acceptable interaction with XR agents.

Best Talk at MultimodalAI 2026. The presentation “UDIVA HHOI: A Multimodal Dataset for Context-Aware Multi-Actor Interaction Understanding” received the meeting's Best Talk recognition.

REACT 2026 challenge paper released. The fourth challenge introduces personalised facial reaction generation and EEG reaction prediction.

Paper published on children's perceived emotions. The study examines facial expressions in children aged three to five while they play with apps.

PhD completion. My PhD student Germán Barquero successfully defended “On the Move: Towards Realistic and Controllable Human Motion Generation” on 30 January.

Royal Academy of Engineering Research Fellowship. Awarded a five-year fellowship to develop ECASIS and advance socially intelligent systems with hybrid AI.

REACT 2025 paper at ACM Multimedia. The third challenge introduced MARS, a large-scale dataset for multiple appropriate facial reaction generation.

Paper accepted at ACII 2025. “Enhancing Personality Recognition by Comparing the Predictive Power of Traits, Facets, and Nuances.”

First PhD completion. Zejian Zhang successfully defended “Deep Learning Approaches for Human Activity Understanding.”

Extraordinary PhD Thesis Award. Recognised by the Department of Mathematics and Informatics at the University of Barcelona.

Two papers accepted at CVPR 2025. Rolling Prediction Models and MixerMDM advance controllable human motion generation.

Paper accepted in IEEE Transactions on Affective Computing. Exploring emotion expression recognition in older adults interacting with a virtual coach.