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: Jun 24, 2025

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K

Human activity recognition using a single-photon direct time-of-flight sensor.

Germán Mora-Martín, Stirling Scholes, Robert K Henderson

    Optics Express
    |June 11, 2024
    PubMed
    Summary

    This study introduces a novel vision system using Single-Photon Avalanche Diode (SPAD) sensors for human activity recognition. The system achieves 89% accuracy in distinguishing seven activities from 40m, even with low resolution.

    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

    Single pixel image classification using an ultrafast digital light projector.

    Optics express·2026
    Same author

    Spatial dealiasing of classical geomagnetic survey data through use of a microfabricated wearable quantum magnetometer.

    Scientific reports·2026
    Same author

    Characterization of a Pixel-level Binarized Autocorrelator for Large Array Multispeckle Diffuse Correlation Spectroscopy under Photon-starved Conditions.

    IEEE transactions on biomedical circuits and systems·2026
    Same author

    Compensating for photon counting losses in a TCSPC SPAD array enables quantitative time-resolved fluorescence anisotropy imaging.

    Methods and applications in fluorescence·2026
    Same author

    Human activity recognition at a kilometer range using single-photon LiDAR.

    Optics express·2026
    Same author

    Underwater 3D imaging using a single-photon avalanche diode detector array with multi-event time-to-digital conversion.

    Optics express·2026

    Area of Science:

    • Computer Vision
    • Sensor Technology
    • Artificial Intelligence

    Background:

    • Single-Photon Avalanche Diode (SPAD) direct Time-of-Flight (dToF) sensors offer long-distance depth imaging capabilities, useful for detecting objects regardless of color or texture.
    • Challenges exist in applying traditional computer vision to SPAD data due to low resolution and solar interference, especially for distant objects represented by few pixels.

    Purpose of the Study:

    • To develop a robust human activity recognition system using SPAD-based depth data.
    • To overcome limitations of low transverse resolution and noise in SPAD sensors for real-world applications.
    • To train the system entirely on synthetic data for broad applicability.

    Main Methods:

    • A novel SPAD-based vision system was designed, integrating convolutional and recurrent neural networks.

    More Related Videos

    A Protocol for Real-time 3D Single Particle Tracking
    10:16

    A Protocol for Real-time 3D Single Particle Tracking

    Published on: January 3, 2018

    14.9K
    Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals
    05:52

    Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals

    Published on: October 20, 2019

    36.1K

    Related Experiment Videos

    Last Updated: Jun 24, 2025

    Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
    06:49

    Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

    Published on: December 11, 2015

    8.9K
    A Protocol for Real-time 3D Single Particle Tracking
    10:16

    A Protocol for Real-time 3D Single Particle Tracking

    Published on: January 3, 2018

    14.9K
    Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals
    05:52

    Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals

    Published on: October 20, 2019

    36.1K
  • The system was trained exclusively on synthetic data to handle real-world sensor limitations.
  • Performance was evaluated using real data from a 64x32 pixel SPAD sensor at distances up to 40m.

    • Main Results:

    • The system achieved an average accuracy of 89% in distinguishing between seven different human activities.
    • It successfully addressed challenges posed by limited transverse resolution (human limbs appearing as ~1 pixel wide) and noise.
    • The system processes continuous video-rate depth data streams at up to 66 FPS on a GPU.

    Conclusions:

    • The proposed SPAD-based vision system demonstrates high accuracy for human activity recognition in challenging, long-distance scenarios.
    • Training on synthetic data proves effective for overcoming real-world sensor limitations.
    • The system's real-time processing capability makes it suitable for surveillance and autonomous systems.