Pia Bideau

Papers

Total Citations

H-Index

About

Pia Bideau’s research sits at the intersection of computer vision, human-robot interaction, and autonomous systems, with a particular focus on understanding and anticipating human behavior in dynamic environments. Her most influential work, “Action-Based Contrastive Learning for Trajectory Prediction” (2022), addresses the critical challenge of predicting pedestrian trajectories from a first-person, moving-camera perspective—a problem central to safe autonomous driving and seamless human-robot collaboration. By introducing a novel action-based contrastive learning framework, Bideau enables models to leverage observed human actions as supervisory signals, improving trajectory forecasts even under complex egomotion. This paper has garnered over 48 citations, reflecting its timely impact on the field. Bideau’s contributions are notable for bridging action recognition and predictive modeling, offering a more intuitive and robust approach to anticipating human intent. Her work is particularly valuable for researchers developing socially-aware autonomous systems, as it provides a principled method for integrating behavioral cues into trajectory prediction pipelines. Through her innovative use of contrastive learning, Bideau has advanced the state of the art in human-robot interaction, making her a rising voice in embodied AI and safe navigation.