One-Shot Open-Set Skeleton-Based Action Recognition

This video demonstrates a novel model for few shot, open set action recognition, i.e., a One-Shot model that is augmented with a discriminator that rejects unknown actions. This model is useful for applications in humanoid robotics, because it allows to easily add new classes and determine whether an input sequence is among the ones that are known to the system.

Details in the paper:
One-Shot Open-Set Skeleton-Based Action Recognition, S. Berti, A. Rosasco, M. Colledanchise, L. Natale