Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Detection of acute lymphocytic and myeloid leukaemia with optimized dimensionality reduction and deep learning classification framework.

Current research in translational medicine·2026
Same author

First aid awareness among non-medical staff of armed forces hospitals in Taif region.

Journal of family medicine and primary care·2026
Same author

Using wearable sensors during the timed up and go and 6-minute walk test in Spinal Bulbar Muscular Atrophy.

PLOS digital health·2026
Same author

Bridging modalities: a deep learning framework for brain tumor classification via CT-MRI integration and model fusion.

Frontiers in computational neuroscience·2026
Same author

A hybrid optimized framework with energy shape prior segmentation for brain tumor detection in MRI images.

Digital health·2026
Same author

A Preliminary Study on Comparative Eye Tracking Analysis Using a Meta Quest Pro.

Inquiry : a journal of medical care organization, provision and financing·2026

Related Experiment Video

Updated: Sep 3, 2025

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT
08:04

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Published on: April 23, 2020

6.9K

Human Gait Analysis: A Sequential Framework of Lightweight Deep Learning and Improved Moth-Flame Optimization

Muhammad Attique Khan1, Habiba Arshad2, Robertas Damaševičius3

  • 1Department of Computer Science, HITEC University, Taxila, Pakistan.

Computational Intelligence and Neuroscience
|July 25, 2022
PubMed
Summary

This study introduces a lightweight deep learning approach for human gait recognition, enhancing accuracy in biometric identification and video surveillance. The method significantly improves performance over traditional techniques.

More Related Videos

Low-Cost Gait Analysis for Behavioral Phenotyping of Mouse Models of Neuromuscular Disease
05:53

Low-Cost Gait Analysis for Behavioral Phenotyping of Mouse Models of Neuromuscular Disease

Published on: July 18, 2019

16.8K
Comprehensive Understanding of Inactivity-Induced Gait Alteration in Rodents
04:37

Comprehensive Understanding of Inactivity-Induced Gait Alteration in Rodents

Published on: July 6, 2022

2.5K

Related Experiment Videos

Last Updated: Sep 3, 2025

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT
08:04

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Published on: April 23, 2020

6.9K
Low-Cost Gait Analysis for Behavioral Phenotyping of Mouse Models of Neuromuscular Disease
05:53

Low-Cost Gait Analysis for Behavioral Phenotyping of Mouse Models of Neuromuscular Disease

Published on: July 18, 2019

16.8K
Comprehensive Understanding of Inactivity-Induced Gait Alteration in Rodents
04:37

Comprehensive Understanding of Inactivity-Induced Gait Alteration in Rodents

Published on: July 6, 2022

2.5K

Area of Science:

  • Computer Science
  • Biometrics
  • Artificial Intelligence

Background:

  • Human gait recognition is crucial for biometrics and video surveillance.
  • Traditional methods suffer from accuracy and time inefficiencies due to feature extraction and classification.
  • Deep learning offers potential for improved performance in these areas.

Purpose of the Study:

  • To propose a lightweight deep learning method for accurate human gait recognition.
  • To address the limitations of traditional feature extraction and classification techniques.
  • To enhance biometric identification and video surveillance capabilities.

Main Methods:

  • Utilized two lightweight, pre-trained deep learning models, fine-tuned with additional layers and frozen middle layers.
  • Employed deep transfer learning for model training and engineered features from fully connected and average pooling layers.
  • Integrated discriminant correlation analysis for feature fusion, optimized with an improved moth-flame optimization algorithm, and classified using an extreme learning machine (ELM).

Main Results:

  • Achieved an average accuracy of 91.20% on the CASIA B dataset and 98.60% on the TUM GAID dataset.
  • Demonstrated superior accuracy compared to existing state-of-the-art methods.
  • Validated the effectiveness of the proposed lightweight deep learning approach.

Conclusions:

  • The proposed lightweight deep learning method significantly enhances human gait recognition accuracy.
  • This approach offers a more efficient and accurate solution for biometric identification and video surveillance applications.
  • The study highlights the potential of optimized deep transfer learning for complex pattern recognition tasks.