An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition

This paper presents a human activity recognition framework tailored for healthcare applications, emphasizing the essential balance between accuracy and interpretability required for medical monitoring. The model utilizes MediaPipe to capture the complex dynamics of human movements and introduce an i...

Full description

Saved in:
Bibliographic Details
Main Authors: Micheal Dutt, Morten Goodwin, Christian W. Omlin
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10606472/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850064889343639552
author Micheal Dutt
Morten Goodwin
Christian W. Omlin
author_facet Micheal Dutt
Morten Goodwin
Christian W. Omlin
author_sort Micheal Dutt
collection DOAJ
description This paper presents a human activity recognition framework tailored for healthcare applications, emphasizing the essential balance between accuracy and interpretability required for medical monitoring. The model utilizes MediaPipe to capture the complex dynamics of human movements and introduce an interpretable feature reduction function. This method improves traditional dimensionality reduction techniques like principal component analysis. Our feature engineering is based on the importance of feature permutations; it selectively retains salient features, thus enhancing the interpretability essential for the medical domain. We validated our method on the “NTU RGB+D” dataset; it improves the recognition accuracy for a range of human activities that may be relevant for elderly care. However, the recognition of subtler activities like neck pain and headaches requires further investigation. This study underscores the potential to advance patient monitoring and sets the stage for its expanded application in various medical contexts.
format Article
id doaj-art-4f7585bbdcb04eecafcdf147492722dc
institution DOAJ
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-4f7585bbdcb04eecafcdf147492722dc2025-08-20T02:49:09ZengIEEEIEEE Access2169-35362024-01-011218794718796310.1109/ACCESS.2024.343277610606472An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity RecognitionMicheal Dutt0https://orcid.org/0000-0001-8800-0977Morten Goodwin1https://orcid.org/0000-0001-6331-702XChristian W. Omlin2https://orcid.org/0000-0003-0299-171XCentre for Artificial Intelligence Research (CAIR), University of Agder, Grimstad, NorwayCentre for Artificial Intelligence Research (CAIR), University of Agder, Grimstad, NorwayCentre for Artificial Intelligence Research (CAIR), University of Agder, Grimstad, NorwayThis paper presents a human activity recognition framework tailored for healthcare applications, emphasizing the essential balance between accuracy and interpretability required for medical monitoring. The model utilizes MediaPipe to capture the complex dynamics of human movements and introduce an interpretable feature reduction function. This method improves traditional dimensionality reduction techniques like principal component analysis. Our feature engineering is based on the importance of feature permutations; it selectively retains salient features, thus enhancing the interpretability essential for the medical domain. We validated our method on the “NTU RGB+D” dataset; it improves the recognition accuracy for a range of human activities that may be relevant for elderly care. However, the recognition of subtler activities like neck pain and headaches requires further investigation. This study underscores the potential to advance patient monitoring and sets the stage for its expanded application in various medical contexts.https://ieeexplore.ieee.org/document/10606472/Artificial intelligencemachine learningdeep learninghuman activity recognitionhealthcarecomputer vision
spellingShingle Micheal Dutt
Morten Goodwin
Christian W. Omlin
An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition
IEEE Access
Artificial intelligence
machine learning
deep learning
human activity recognition
healthcare
computer vision
title An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition
title_full An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition
title_fullStr An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition
title_full_unstemmed An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition
title_short An Interpretable Deep Learning-Based Feature Reduction in Video-Based Human Activity Recognition
title_sort interpretable deep learning based feature reduction in video based human activity recognition
topic Artificial intelligence
machine learning
deep learning
human activity recognition
healthcare
computer vision
url https://ieeexplore.ieee.org/document/10606472/
work_keys_str_mv AT michealdutt aninterpretabledeeplearningbasedfeaturereductioninvideobasedhumanactivityrecognition
AT mortengoodwin aninterpretabledeeplearningbasedfeaturereductioninvideobasedhumanactivityrecognition
AT christianwomlin aninterpretabledeeplearningbasedfeaturereductioninvideobasedhumanactivityrecognition
AT michealdutt interpretabledeeplearningbasedfeaturereductioninvideobasedhumanactivityrecognition
AT mortengoodwin interpretabledeeplearningbasedfeaturereductioninvideobasedhumanactivityrecognition
AT christianwomlin interpretabledeeplearningbasedfeaturereductioninvideobasedhumanactivityrecognition