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

A hybrid approach for diabetic retinopathy stages classification using spatial and textural features.

Health informatics journal·2026
Same author

Deep Learning Based Framework for Detection and Classification of Leukemia Using Microscopic Images.

Microscopy research and technique·2026
Same author

Club cells in asthma: Multifunctional regulators of airway integrity and emerging therapeutic targets.

Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie·2026
Same author

Emergence and genomic characterization of Listeria monocytogenes causing human listeriosis in Pakistan, 2017-2023.

BMC microbiology·2026
Same author

Emerging role of SIRT1 in asthma and COPD from molecular mechanisms to translational therapy.

iScience·2026
Same author

Methicillin-Resistant <i>Staphylococcus aureus</i> breast abscesses: Risk factors and outcomes from a tertiary care center of a lower-middle-income country.

Pakistan journal of medical sciences·2026

Related Experiment Video

Updated: Sep 15, 2025

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K

Content oriented 3D-CNN sequence learning architecture for academic activities recognition using a realistic CAD

Muhammad Wasim1, Imran Ahmed2, Naveed Abbas1

  • 1Department of Computer Science, Islamia College Peshawar, Peshawar, Pakistan.

Scientific Reports
|July 12, 2025
PubMed
Summary
This summary is machine-generated.

Researchers developed a lightweight 3D-CNN model for academic activity recognition using campus videos. This model achieves 95% accuracy with low computational cost, outperforming LSTM for efficient video analysis.

Keywords:
3D CNNActivity recognitionDeep learningLong short-term memoryRNNSequence learningVideo surveillance

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.4K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

Related Experiment Videos

Last Updated: Sep 15, 2025

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.4K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Machine Learning

Background:

  • Human activity recognition is crucial in computer vision.
  • Academic institutions possess extensive video data from campus surveillance.
  • There is a need for efficient models to recognize academic activities.

Purpose of the Study:

  • To propose a lightweight 3D-CNN architecture for academic activity recognition.
  • To utilize spatial and temporal video information for sequence learning.
  • To evaluate the model's performance against state-of-the-art algorithms.

Main Methods:

  • Development of a novel lightweight 3D-CNN architecture.
  • Training and testing on a realistic campus video dataset.
  • Comparative analysis with the Long Short-Term Memory (LSTM) model.

Main Results:

  • The proposed 3D-CNN model achieved 95% accuracy.
  • Demonstrated a computational cost of 13.3 GFLOPs.
  • Exhibited a low memory overhead of 18,464 KB.

Conclusions:

  • The lightweight 3D-CNN model is highly effective for academic activity recognition.
  • The model offers superior performance compared to LSTM.
  • Its efficiency makes it suitable for real-world campus surveillance applications.