Can Ensemble of Classifiers Provide Better Recognition Results in Packaging Activity?

TL;DR

This paper proposes an ensemble machine learning approach for human activity recognition using skeleton-based MoCap data, achieving high accuracy on a packaging activity dataset despite limited data.

Contribution

It introduces a novel ensemble methodology tailored for MoCap datasets, improving recognition accuracy in industrial activity contexts.

Findings

Achieved 98% accuracy with 10-fold Cross-Validation

Achieved 82% accuracy with Leave-One-Out-Cross-Validation

Demonstrated effectiveness on Bento Packaging Activity dataset

Abstract

Skeleton-based Motion Capture (MoCap) systems have been widely used in the game and film industry for mimicking complex human actions for a long time. MoCap data has also proved its effectiveness in human activity recognition tasks. However, it is a quite challenging task for smaller datasets. The lack of such data for industrial activities further adds to the difficulties. In this work, we have proposed an ensemble-based machine learning methodology that is targeted to work better on MoCap datasets. The experiments have been performed on the MoCap data given in the Bento Packaging Activity Recognition Challenge 2021. Bento is a Japanese word that resembles lunch-box. Upon processing the raw MoCap data at first, we have achieved an astonishing accuracy of 98% on 10-fold Cross-Validation and 82% on Leave-One-Out-Cross-Validation by using the proposed ensemble model.