Understanding Human Manipulation With the Environment: A Novel Taxonomy for Video Labelling

In recent years, the spread of data-driven approaches for robotic grasp synthesis has come with increasing need reliable datasets, which can be built e.g. through video labelling. To this goal, it is important to define suitable rules characterize main human types, easily identifying them in streams. work, we present a novel taxonomy that builds upon related state art, but specifically thought It focuses on interaction hand environment and accounts pre-contact phases, bi-manual grasps as well non-prehensile strategies. This study complemented dataset labelled videos subjects performing activities daily living, total nine hours, description MatLab tools labelling new videos. Both hands were at any time. We used these data preliminary statistical occurrences here proposed class types.