Smart Arse: Posture Classification with Textile Sensors in Trousers

“We talk ‘through’ our clothes, if we want [sic] or not.”

– say a research team from Queen Mary University of London, who have recently created ‘Smart Arse’ – a pair of ‘sensing trousers’ (that’s ‘sensing pants’ US) – which can reveal details of the ‘conversational engagement’ of the wearers.

“People continually adjust their posture and re-arrange their hands and legs when seated. For example: hands resting on laps; elbows on thighs; a forward leaning posture; hands that are tucked between thighs; hands on knees and many other variations. Our basic question is, what can these different postural states tell us about conversational engagement and potentially even about more complex, affective states?”

In Smart Arse, an electronic textile pressure matrix around the thighs and buttocks is linked to a machine-learning model which is able to distinguish between 19 posture types, shown promising results, they say, towards an objective to make garments “socially aware”.

See: Smart Arse: Posture Classification with Textile Sensors in Trousers  in ICMI ’18 Proceedings of the 20th ACM International Conference on Multimodal Interaction, Pages 116-124