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 Experiment Video

Updated: Jan 9, 2026

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring
08:16

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

Published on: October 24, 2025

434

Segmentation-Guided Denoising and Gradient-Aware Depth Completion for Enhanced LiDAR-Based Human Activity

Maxime Chebib, Mondher Bouazizi, Tomoaki Otsuki

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
    |December 3, 2025
    PubMed
    Summary
    This summary is machine-generated.

    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

    Non-contact Heart Rate Estimation Using VMD with Spectrogram-Based Preprocessing via CW Radar.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same author

    Stress Detection Using Eye Activity via Recurrent Neural Networks.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same author

    Deep Learning-Based Human Joint Localization Using mmWave Radar and Sequential Frame Fusion.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same author

    Beyond Contact: An Open-Set Biometric Identification System Using Radar-Extracted Heart Signals.

    IEEE journal of biomedical and health informatics·2025
    Same author

    A Novel Deep Learning Model for Human Skeleton Estimation Using FMCW Radar.

    Sensors (Basel, Switzerland)·2025
    Same author

    Deep Learning-based Open-set Person Identification using Radar Extracted Cardiac Signals.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same journal

    Analysis of End-Tidal CO2 Variability During Plateau Waves Episodes: An Information Theoretic Approach<sup></sup>.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same journal

    AI and Tomosynthesis for Breast Cancer Molecular Subtyping: A step toward precision medicine<sup></sup>.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same journal

    Towards Sustainable Protein Recovery from Biological Waste: Assessing Polyethersulfone-based Microfiltration.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same journal

    Analysis of the cardiovascular response to standardized polymicrobial peritonitis experimental model.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same journal

    Automated Wrist Ultrasound Image Bone Enhancement and Segmentation Using Deep Learning.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    Same journal

    A Deep Learning approach for Depressive Symptoms assessment in Parkinson's disease patients using facial videos.

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
    See all related articles

    This study introduces a new method to improve 3D-LiDAR depth images for Human Activity Recognition (HAR). It uses segmentation and gradient-guided completion to enhance depth data without needing ground-truth LiDAR.

    Area of Science:

    • Robotics and Computer Vision
    • Sensor Data Processing
    • Human-Computer Interaction

    Background:

    • 3D-LiDAR sensors face challenges in Human Activity Recognition (HAR) due to occlusions, noise, and sparse data.
    • Existing methods often require difficult-to-obtain ground truth LiDAR datasets and are primarily for autonomous driving.
    • There is a need for efficient and accurate LiDAR data enhancement specifically for HAR applications.

    Purpose of the Study:

    • To develop a novel approach for enhancing 3D-LiDAR depth images for Human Activity Recognition (HAR).
    • To address challenges of noise, sparsity, and occlusions in LiDAR data for HAR.
    • To create a method that does not rely on ground-truth LiDAR datasets.

    Main Methods:

    • A U-Net-based human segmentation model is employed to remove background noise and irrelevant data.

    More Related Videos

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    996

    Related Experiment Videos

    Last Updated: Jan 9, 2026

    Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring
    08:16

    Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

    Published on: October 24, 2025

    434
    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    996
  • A gradient-guided depth completion algorithm is utilized to interpolate missing depth values, preserving structural integrity.
  • RGB and depth data are calibrated to create aligned training datasets for the segmentation model.

    • Main Results:

    • The segmentation model achieved high accuracy (Dice Score: 0.958 on seen, 0.923 on unseen subjects).
    • The proposed method significantly improves LiDAR depth image quality.
    • The approach maintains computational efficiency, suitable for real-time HAR.

    Conclusions:

    • This work presents the first approach combining segmentation-based denoising with gradient-aware depth completion for LiDAR-based HAR.
    • The method effectively enhances LiDAR depth data for HAR without requiring ground-truth LiDAR datasets.
    • The developed technique offers a computationally efficient solution for real-time HAR applications.