Human-centred Machine Perception

The ability to automatically perceive and interpret human behavior is crucial for adaptive intelligent technology (e.g., social robots or tutoring systems). Not only can it improve user understanding in applications, but it also holds the potential for novel scientific insights about human behavior in the real-world. This course covers relevant theories and principles for developing Human-centred Machine Perception systems that leverage sensor data (e.g., from cameras, microphones, or wearables) in individual, dyadic, and group settings. Topics covered include: data collection, annotation, implementation, and evaluation of such systems, as well as relevant theories about human behavior and psychology (e.g., vocal expression, group dynamics, and human emotion).