A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition

Human activity recognition (HAR) is a challenging computer vision problem that requires recognizing and categorizing human actions using spatiotemporal data. In recent years, ConvLSTM has shown distinctive advances in manipulating spatiotemporal data. ConvLSTM-based architectures, as any deep learni...

Full description

Saved in:
Bibliographic Details
Main Authors: Sarah Khater, Magda B. Fayek, Mayada Hadhoud
Format: Article
Language:English
Published: Graz University of Technology 2025-04-01
Series:Journal of Universal Computer Science
Subjects:
Online Access:https://lib.jucs.org/article/131543/download/pdf/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850281358832697344
author Sarah Khater
Magda B. Fayek
Mayada Hadhoud
author_facet Sarah Khater
Magda B. Fayek
Mayada Hadhoud
author_sort Sarah Khater
collection DOAJ
description Human activity recognition (HAR) is a challenging computer vision problem that requires recognizing and categorizing human actions using spatiotemporal data. In recent years, ConvLSTM has shown distinctive advances in manipulating spatiotemporal data. ConvLSTM-based architectures, as any deep learning architecture, require deciding on many hyperparameters apart from trainable weights. State-of-the-art designs for general purpose datasets already exist, but specific purpose applications require architecture designs that perform well on application-dependent datasets. The design of such architectures requires either many trials and errors, which consume time and resources, or an experienced architect. Neural architecture search (NAS) meth-ods have been introduced to automate the design process and address the challenge of relying on expert knowledge when creating neural architectures. NAS enables rapid prototyping and experimentation, reducing the time spent on trial and error in manual design. One of the leading approaches in NAS is Genetic Algorithm (GA), which plays a significant role in optimizing neu-ral architectures. In this paper, a novel GA-based approach is proposed to automatically design ConvLSTM-based architectures from scratch for HAR applications. Our approach is based on multi-objective GA that maximizes recognition accuracy and minimizes the number of trainable parameters and overfitting measure. The experiments are held on KTH, Weizmann, and UCF Sports datasets. The best classification accuracies from the generated models are 97.92%, 96.77%, and 94.87% for KTH, Weizmann, and UCF Sports datasets, respectively. The experimental results show that the automatically generated models with the proposed approach outperform some of the state-of-the-art manually designed ConvLSTM-based architectures with percentages up to 9.92%, 5.77% and 23.64% for KTH, Weizmann, and UCF Sports, respectively. We also compared our approach with other NAS approaches. Our approach is found to outperform some of the introduced approaches with percentages approximately 2%, 11%, and 4% for KTH, Weizmann, and UCF Sports, respectively.
format Article
id doaj-art-67d0d00c5a6f4fd7a25fc4a3fafbc091
institution OA Journals
issn 0948-6968
language English
publishDate 2025-04-01
publisher Graz University of Technology
record_format Article
series Journal of Universal Computer Science
spelling doaj-art-67d0d00c5a6f4fd7a25fc4a3fafbc0912025-08-20T01:48:20ZengGraz University of TechnologyJournal of Universal Computer Science0948-69682025-04-0131549451810.3897/jucs.131543131543A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity RecognitionSarah Khater0Magda B. Fayek1Mayada Hadhoud2Cairo UniversityCairo UniversityCairo UniversityHuman activity recognition (HAR) is a challenging computer vision problem that requires recognizing and categorizing human actions using spatiotemporal data. In recent years, ConvLSTM has shown distinctive advances in manipulating spatiotemporal data. ConvLSTM-based architectures, as any deep learning architecture, require deciding on many hyperparameters apart from trainable weights. State-of-the-art designs for general purpose datasets already exist, but specific purpose applications require architecture designs that perform well on application-dependent datasets. The design of such architectures requires either many trials and errors, which consume time and resources, or an experienced architect. Neural architecture search (NAS) meth-ods have been introduced to automate the design process and address the challenge of relying on expert knowledge when creating neural architectures. NAS enables rapid prototyping and experimentation, reducing the time spent on trial and error in manual design. One of the leading approaches in NAS is Genetic Algorithm (GA), which plays a significant role in optimizing neu-ral architectures. In this paper, a novel GA-based approach is proposed to automatically design ConvLSTM-based architectures from scratch for HAR applications. Our approach is based on multi-objective GA that maximizes recognition accuracy and minimizes the number of trainable parameters and overfitting measure. The experiments are held on KTH, Weizmann, and UCF Sports datasets. The best classification accuracies from the generated models are 97.92%, 96.77%, and 94.87% for KTH, Weizmann, and UCF Sports datasets, respectively. The experimental results show that the automatically generated models with the proposed approach outperform some of the state-of-the-art manually designed ConvLSTM-based architectures with percentages up to 9.92%, 5.77% and 23.64% for KTH, Weizmann, and UCF Sports, respectively. We also compared our approach with other NAS approaches. Our approach is found to outperform some of the introduced approaches with percentages approximately 2%, 11%, and 4% for KTH, Weizmann, and UCF Sports, respectively.https://lib.jucs.org/article/131543/download/pdf/ConvLSTMHARKTHMulti-objective fitnessNAS
spellingShingle Sarah Khater
Magda B. Fayek
Mayada Hadhoud
A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition
Journal of Universal Computer Science
ConvLSTM
HAR
KTH
Multi-objective fitness
NAS
title A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition
title_full A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition
title_fullStr A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition
title_full_unstemmed A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition
title_short A Novel GA-based Approach to Automatically Generate ConvLSTM Architectures for Human Activity Recognition
title_sort novel ga based approach to automatically generate convlstm architectures for human activity recognition
topic ConvLSTM
HAR
KTH
Multi-objective fitness
NAS
url https://lib.jucs.org/article/131543/download/pdf/
work_keys_str_mv AT sarahkhater anovelgabasedapproachtoautomaticallygenerateconvlstmarchitecturesforhumanactivityrecognition
AT magdabfayek anovelgabasedapproachtoautomaticallygenerateconvlstmarchitecturesforhumanactivityrecognition
AT mayadahadhoud anovelgabasedapproachtoautomaticallygenerateconvlstmarchitecturesforhumanactivityrecognition
AT sarahkhater novelgabasedapproachtoautomaticallygenerateconvlstmarchitecturesforhumanactivityrecognition
AT magdabfayek novelgabasedapproachtoautomaticallygenerateconvlstmarchitecturesforhumanactivityrecognition
AT mayadahadhoud novelgabasedapproachtoautomaticallygenerateconvlstmarchitecturesforhumanactivityrecognition