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

Label-Free Prediction of EGFR Mutation Status Using Fluorescence Lifetime Imaging and Deep Learning in Lung Adenocarcinoma.

Cancer research·2026
Same author

Phthalazine-based quaternary ammonium salts: synthesis, biological evaluation and membrane-targeting mechanism against <i>Staphylococcus aureus</i>.

Frontiers in microbiology·2026
Same author

Exposure to multiple metallic elements and risk of thyroid tumors: insights from elemental profiling, diet, and molecular characteristics plasma levels of metallic elements.

Frontiers in oncology·2026
Same author

Novel Perspective for Prognostic Stratification and Personalized Therapy in Breast Cancer Patients: Development of Cancer Stem Cells and Metabolism-Associated Prognostic Model.

International journal of women's health·2026
Same author

Rare <i>BRCA1</i> c.3418_3419insTGACTACT:p.S1140Mfs*18 germline mutation in a family with breast and ovarian cancer.

Oncology letters·2026
Same author

Discovery of Potent 1,3-Cyclobutane-Containing Dual A<sub>2A</sub>/A<sub>2B</sub> Receptor Antagonists with Low Projected Human Dose for the Treatment of Cancer.

ACS medicinal chemistry letters·2026

Related Experiment Video

Updated: Oct 28, 2025

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

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.5K

Non-fusion time-resolved depth image reconstruction using a highly efficient neural network architecture.

Zhenya Zang, Dong Xiao, David Day-Uei Li

    Optics Express
    |July 16, 2021
    PubMed
    Summary

    This study introduces a novel neural network for reconstructing depth images from noisy single-photon avalanche diode (SPAD) sensor data. The efficient, quantized design enhances 3D LiDAR performance in low light.

    More Related Videos

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
    12:27

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

    Published on: February 15, 2017

    7.1K
    Deep Learning-Based Segmentation of Cryo-Electron Tomograms
    10:25

    Deep Learning-Based Segmentation of Cryo-Electron Tomograms

    Published on: November 11, 2022

    9.7K

    Related Experiment Videos

    Last Updated: Oct 28, 2025

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

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    9.5K
    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
    12:27

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

    Published on: February 15, 2017

    7.1K
    Deep Learning-Based Segmentation of Cryo-Electron Tomograms
    10:25

    Deep Learning-Based Segmentation of Cryo-Electron Tomograms

    Published on: November 11, 2022

    9.7K

    Area of Science:

    • Photonics and Sensor Technology
    • Computer Vision and Machine Learning
    • Robotics and Autonomous Systems

    Background:

    • Single-photon avalanche diodes (SPADs) offer high sensitivity for 3D light detection and ranging (LiDAR) in low-light conditions.
    • Accurate depth map reconstruction from noisy time-of-arrival (ToA) data generated by SPADs presents a significant challenge.
    • Existing methods often require fusion with other data sources or are computationally intensive for embedded systems.

    Purpose of the Study:

    • To develop a photon-efficient neural network for direct depth image reconstruction from SPAD ToA data.
    • To compress the neural network using low-bit quantization for embedded hardware compatibility.
    • To achieve high-fidelity depth reconstruction with improved accuracy and reduced model complexity.

    Main Methods:

    • Proposed a novel, non-fusion neural network architecture for direct depth image reconstruction.
    • Implemented a low-bit quantization scheme to compress the neural network model.
    • Evaluated the network's performance on time-of-arrival (ToA) point cloud data from SPAD sensors.

    Main Results:

    • The proposed neural network directly reconstructs high-fidelity depth images from ToA data without auxiliary inputs.
    • The quantized network achieves superior reconstruction accuracy compared to existing methods.
    • The compressed architecture significantly reduces the number of parameters, enabling embedded deployment.

    Conclusions:

    • The developed quantized neural network offers an effective solution for accurate depth estimation from SPAD-based LiDAR systems.
    • This approach enhances the feasibility of deploying advanced 3D sensing on resource-constrained embedded platforms.
    • The photon-efficient and accurate reconstruction capabilities open new avenues for low-light 3D imaging applications.