Research topics

We combine deep learning, speech/signal processing and computer vision to develop human-centric AI technologies for domains of direct societal importance. Our group is currently working on the following topics:

• Affect and mental health: Emotion from verbal and non-verbal cues, anxiety and depression detection, monitoring and intervention for wellbeing

• Privacy and bias: Privacy risks and mitigation in speech AI models for affect and behaviour analysis, bias in facial expression perception, diversity & inclusion in screen media.

• Activity and interaction: Driver activity monitoring for L2+, pedestrian and crowd monitoring for AV and surveillance, action forecasting, human-human interaction modeling for human-robot interactions.