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

Computed Tomography01:10

Computed Tomography

9.2K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
9.2K
Parallel Processing01:20

Parallel Processing

840
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
840

You might also read

Related Articles

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

Sort by
Same author

Fully-connected microwave photonic multi-beamformer with fast beam-steering for broadband wireless communication.

Nature communications·2026
Same author

Integrated bionic LiDAR for adaptive 4D machine vision.

Nature communications·2025
Same author

A 4×256 Gbps silicon transmitter with on-chip adaptive dispersion compensation.

Nature communications·2025
Same author

Parallel wavelength-division-multiplexed signal transmission and dispersion compensation enabled by soliton microcombs and microrings.

Nature communications·2024
Same author

Broadband continuously tunable microwave photonic delay line based on cascaded silicon microrings.

Optics express·2021
Same author

Comparison of the phase change process in a GST-loaded silicon waveguide and MMI.

Optics express·2021
Same journal

Two-photon 3D imaging of optically stimulated neural activity at 100 Hz.

Light, science & applications·2026
Same journal

Quasi-bound states in the continuum driven photoresponse in multiple quantum wells for machine vision.

Light, science & applications·2026
Same journal

Spin-photon qubits for scalable quantum network.

Light, science & applications·2026
Same journal

Dual-mode switchable and reconfigurable Van der Waals phototransistor for multi-state image encryption.

Light, science & applications·2026
Same journal

Weak polarization electric field Ⅲ-N LEDs on polar plane with enhanced efficiency and strong lateral carrier confinement.

Light, science & applications·2026
Same journal

Bi-layer photonic random meta-composite for cryogenic thermal control by ultra-broadband scattering matched reflectance.

Light, science & applications·2026
See all related articles

Related Experiment Video

Updated: Mar 8, 2026

Three-dimensional Particle Tracking Velocimetry for Turbulence Applications: Case of a Jet Flow
13:02

Three-dimensional Particle Tracking Velocimetry for Turbulence Applications: Case of a Jet Flow

Published on: February 27, 2016

13.1K

Integrated photonic 3D tensor processing engine.

Yue Wu1, Ziheng Ni1, Xin Li1

  • 1State Key Laboratory of Photonics and Communications, School of Integrated Circuits, Shanghai Jiao Tong University, Shanghai, 200240, China.

Light, Science & Applications
|March 6, 2026
PubMed
Summary
This summary is machine-generated.

This study introduces an integrated photonic 3D tensor processing engine (3D-TPE) that accelerates deep learning by performing computations entirely in the optical domain. This novel approach reduces overheads and enhances efficiency for 3D convolutional neural networks.

More Related Videos

Integrated Photoacoustic Ophthalmoscopy and Spectral-domain Optical Coherence Tomography
11:21

Integrated Photoacoustic Ophthalmoscopy and Spectral-domain Optical Coherence Tomography

Published on: January 15, 2013

12.0K
Uncovering Hidden Dynamics of Natural Photonic Structures Using Holographic Imaging
05:45

Uncovering Hidden Dynamics of Natural Photonic Structures Using Holographic Imaging

Published on: March 31, 2022

3.1K

Related Experiment Videos

Last Updated: Mar 8, 2026

Three-dimensional Particle Tracking Velocimetry for Turbulence Applications: Case of a Jet Flow
13:02

Three-dimensional Particle Tracking Velocimetry for Turbulence Applications: Case of a Jet Flow

Published on: February 27, 2016

13.1K
Integrated Photoacoustic Ophthalmoscopy and Spectral-domain Optical Coherence Tomography
11:21

Integrated Photoacoustic Ophthalmoscopy and Spectral-domain Optical Coherence Tomography

Published on: January 15, 2013

12.0K
Uncovering Hidden Dynamics of Natural Photonic Structures Using Holographic Imaging
05:45

Uncovering Hidden Dynamics of Natural Photonic Structures Using Holographic Imaging

Published on: March 31, 2022

3.1K

Area of Science:

  • Photonics
  • Optical Computing
  • Deep Learning Acceleration

Background:

  • Mainstream photonic accelerators are limited to 2D matrix-vector multiplications, requiring complex data reshaping for 3D tasks.
  • Implementing 3D convolutional neural networks on current hardware incurs significant memory and time overheads due to electrical domain processing and synchronization.
  • External electronic clocks increase system complexity for channel synchronization in photonic accelerators.

Purpose of the Study:

  • To propose and demonstrate an integrated photonic 3D tensor processing engine (3D-TPE) for efficient 3D convolutional neural network computation.
  • To enable optical domain data caching, channel synchronization, and computation, thereby reducing memory usage and system complexity.
  • To advance optical computing solutions for high-order tensor operations crucial for emerging AI applications.

Main Methods:

  • Developed a 3D-TPE utilizing interleaving modulation of time, wavelength, and space.
  • Integrated an optical tunable delay line (OTDL) chip for optical caching and synchronization up to 200 GHz.
  • Employed a dual-coupled micro-ring resonators (MRRs) based crossbar chip for optical computation.

Main Results:

  • Demonstrated the 3D-TPE's processing capabilities at clock frequencies from 10 GHz to 30 GHz.
  • Achieved 97.06% recognition accuracy in a proof-of-concept LiDAR 3D point cloud image recognition task at 20 GHz.
  • Showcased the potential for optical domain operations, reducing memory and time overheads compared to traditional methods.

Conclusions:

  • The proposed 3D-TPE offers a simplified and efficient solution for high-order tensor convolutions in the optical domain.
  • This technology has significant implications for accelerating AI applications in fields such as autonomous driving, healthcare, and virtual reality.
  • The integrated photonic approach paves the way for next-generation hardware accelerators for complex deep learning models.