Context Action Recognition is curcial for robots to perfoma around humans. Indeed, robot need to asses human action and intentions in order to assist them in everyday life tasks and to collaborative efficiently.
The field of action recognition has aimed to use typical sensors found on robots to recognize agnts, objects and actions they are performing. Typical approach is to record a dataset of various action and label them. But often theses actions are not natural and it can be difficult to represent the variety of ways to perform actions with a lab built dataset.
Context The field of social human-robot interaction is growing. Understanding how communication between humans (human-human) generalises during human-robot communication is crucial in building fluent and enjoyable interactions with social robots. Everyday, more datasets that feature social interaction between humans and between humans and robots are made freely available online.
In this project we propose to take a data-driven approach to build predictive models of social interactions between humans (HH) ( and between humans and robots (HR) interaction using 3 different datasets.