One-Shot Open-Set Skeleton-Based Action Recognition

TL;DR

This paper introduces a novel one-shot deep learning model with a discriminator for open-set skeleton-based action recognition, enabling robots to recognize new actions and reject unknown ones effectively.

Contribution

It proposes an end-to-end trainable model that combines one-shot learning with open-set recognition for action sequences, addressing limitations of existing static image solutions.

Findings

Effective rejection of unknown actions in real-world tests

Ability to add new action classes with minimal data

Superior performance over baseline models in open-set scenarios

Abstract

Action recognition is a fundamental capability for humanoid robots to interact and cooperate with humans. This application requires the action recognition system to be designed so that new actions can be easily added, while unknown actions are identified and ignored. In recent years, deep-learning approaches represented the principal solution to the Action Recognition problem. However, most models often require a large dataset of manually-labeled samples. In this work we target One-Shot deep-learning models, because they can deal with just a single instance for class. Unfortunately, One-Shot models assume that, at inference time, the action to recognize falls into the support set and they fail when the action lies outside the support set. Few-Shot Open-Set Recognition (FSOSR) solutions attempt to address that flaw, but current solutions consider only static images and not sequences of images. Static images remain insufficient to discriminate actions such as sitting-down and standing-up. In this paper we propose a novel model that addresses the FSOSR problem with 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. We show how to train the whole model in an end-to-end fashion and we perform quantitative and qualitative analyses. Finally, we provide real-world examples.