Biomechanically consistent real-time action recognition for human-robot interaction

TL;DR

This paper introduces a real-time human action recognition framework using biomechanical priors and a Transformer-based network, demonstrating high accuracy and robustness in industrial human-robot interaction scenarios.

Contribution

It presents a novel real-time action recognition pipeline that leverages biomechanical joint angles, improving robustness and generalization over existing methods.

Findings

Achieves 88% accuracy in action recognition

Outperforms baseline models in real-time scenarios

Demonstrates effective human-robot interaction in experiments

Abstract

This paper presents a novel framework for real-time human action recognition in industrial contexts, using standard 2D cameras. We introduce a complete pipeline for robust and real-time estimation of human joint kinematics, input to a temporally smoothed Transformer-based network, for action recognition. We rely on a new dataset including 11 subjects performing various actions, to evaluate our approach. Unlike most of the literature that relies on joint center positions (JCP) and is offline, ours uses biomechanical prior, eg. joint angles, for fast and robust real-time recognition. Besides, joint angles make the proposed method agnostic to sensor and subject poses as well as to anthropometric differences, and ensure robustness across environments and subjects. Our proposed learning model outperforms the best baseline model, running also in real-time, along various metrics. It achieves 88% accuracy and shows great generalization ability, for subjects not facing the cameras. Finally, we demonstrate the robustness and usefulness of our technique, through an online interaction experiment, with a simulated robot controlled in real-time via the recognized actions.