Multistage fall detection framework via 3D pose sequences and TCN integration

Abstract An accurate yet computationally efficient fall detection system for sports activities is a significant and challenging task. To address this, we propose a novel multi-stage fall detection framework that integrates 3D pose sequences with temporal convolutional modeling. The framework first p...

Full description

Saved in:
Bibliographic Details
Main Authors: Leitao Qi, Haibo Sun
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-11325-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333566120394752
author Leitao Qi
Haibo Sun
author_facet Leitao Qi
Haibo Sun
author_sort Leitao Qi
collection DOAJ
description Abstract An accurate yet computationally efficient fall detection system for sports activities is a significant and challenging task. To address this, we propose a novel multi-stage fall detection framework that integrates 3D pose sequences with temporal convolutional modeling. The framework first performs 2D human pose estimation to extract and enhance multi-scale spatial features. Then, it reconstructs the 2D poses into 3D poses using a domain transfer architecture that aligns the 2D and 3D poses within a shared semantic space. Subsequently, we introduce a robust fall detection network that leverages temporal convolutions to process the 3D pose sequences, capturing long-term dependencies while maintaining low computational costs for fall event recognition. Evaluated on the large-scale benchmark action dataset NTU RGB+D, our method achieves a fall detection accuracy of 99.87%, demonstrating its state-of-the-art performance and effectiveness.
format Article
id doaj-art-c2dee48390774584b24e0eb4b5351dfa
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-c2dee48390774584b24e0eb4b5351dfa2025-08-20T03:45:48ZengNature PortfolioScientific Reports2045-23222025-07-0115111110.1038/s41598-025-11325-yMultistage fall detection framework via 3D pose sequences and TCN integrationLeitao Qi0Haibo Sun1School of Basic Medical Sciences, Shandong Second Medical UniversitySchool of Basic Medical Sciences, Shandong Second Medical UniversityAbstract An accurate yet computationally efficient fall detection system for sports activities is a significant and challenging task. To address this, we propose a novel multi-stage fall detection framework that integrates 3D pose sequences with temporal convolutional modeling. The framework first performs 2D human pose estimation to extract and enhance multi-scale spatial features. Then, it reconstructs the 2D poses into 3D poses using a domain transfer architecture that aligns the 2D and 3D poses within a shared semantic space. Subsequently, we introduce a robust fall detection network that leverages temporal convolutions to process the 3D pose sequences, capturing long-term dependencies while maintaining low computational costs for fall event recognition. Evaluated on the large-scale benchmark action dataset NTU RGB+D, our method achieves a fall detection accuracy of 99.87%, demonstrating its state-of-the-art performance and effectiveness.https://doi.org/10.1038/s41598-025-11325-yFall detectionHuman pose estimation3D pose liftingDomain transferTemporal convolutional networksAnomaly detection
spellingShingle Leitao Qi
Haibo Sun
Multistage fall detection framework via 3D pose sequences and TCN integration
Scientific Reports
Fall detection
Human pose estimation
3D pose lifting
Domain transfer
Temporal convolutional networks
Anomaly detection
title Multistage fall detection framework via 3D pose sequences and TCN integration
title_full Multistage fall detection framework via 3D pose sequences and TCN integration
title_fullStr Multistage fall detection framework via 3D pose sequences and TCN integration
title_full_unstemmed Multistage fall detection framework via 3D pose sequences and TCN integration
title_short Multistage fall detection framework via 3D pose sequences and TCN integration
title_sort multistage fall detection framework via 3d pose sequences and tcn integration
topic Fall detection
Human pose estimation
3D pose lifting
Domain transfer
Temporal convolutional networks
Anomaly detection
url https://doi.org/10.1038/s41598-025-11325-y
work_keys_str_mv AT leitaoqi multistagefalldetectionframeworkvia3dposesequencesandtcnintegration
AT haibosun multistagefalldetectionframeworkvia3dposesequencesandtcnintegration