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

Parallel Processing01:20

Parallel Processing

186
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...
186

You might also read

Related Articles

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

Sort by
Same author

Cavity solitons as a nonlinear substrate for photonic neuromorphic computing.

Optics express·2026
Same author

Equilibrium propagation for learning in Lagrangian dynamical systems.

Physical review. E·2025
Same author

Weak Kerr nonlinearity boosts the performance of frequency-multiplexed photonic extreme learning machines: a multifaceted approach.

Optics express·2025
Same author

Efficient optimisation of physical reservoir computers using only a delayed input.

Communications engineering·2025
Same author

Deep Photonic Reservoir Computer for Speech Recognition.

IEEE transactions on neural networks and learning systems·2024
Same author

Integrated programmable spectral filter for frequency-multiplexed neuromorphic computers.

Optics express·2023

Related Experiment Video

Updated: Jul 25, 2025

High-resolution, High-speed, Three-dimensional Video Imaging with Digital Fringe Projection Techniques
11:34

High-resolution, High-speed, Three-dimensional Video Imaging with Digital Fringe Projection Techniques

Published on: December 3, 2013

15.7K

High speed human action recognition using a photonic reservoir computer.

Enrico Picco1, Piotr Antonik2, Serge Massar1

  • 1Laboratoire d'Information Quantique, CP 224, Université Libre de Bruxelles (ULB), B-1050, Bruxelles, Belgium.

Neural Networks : the Official Journal of the International Neural Network Society
|June 26, 2023
PubMed
Summary

This study introduces a novel reservoir computing training method for fast and accurate human action recognition in videos. This approach enables real-time processing of multiple video streams, advancing dedicated video processing hardware.

Keywords:
Computer visionHuman action recognitionPhotonicsReservoir computing

More Related Videos

Photodiode-Based Optical Imaging for Recording Network Dynamics with Single-Neuron Resolution in Non-Transgenic Invertebrates
10:18

Photodiode-Based Optical Imaging for Recording Network Dynamics with Single-Neuron Resolution in Non-Transgenic Invertebrates

Published on: July 9, 2020

3.0K
Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories
07:52

Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories

Published on: July 10, 2019

14.3K

Related Experiment Videos

Last Updated: Jul 25, 2025

High-resolution, High-speed, Three-dimensional Video Imaging with Digital Fringe Projection Techniques
11:34

High-resolution, High-speed, Three-dimensional Video Imaging with Digital Fringe Projection Techniques

Published on: December 3, 2013

15.7K
Photodiode-Based Optical Imaging for Recording Network Dynamics with Single-Neuron Resolution in Non-Transgenic Invertebrates
10:18

Photodiode-Based Optical Imaging for Recording Network Dynamics with Single-Neuron Resolution in Non-Transgenic Invertebrates

Published on: July 9, 2020

3.0K
Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories
07:52

Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories

Published on: July 10, 2019

14.3K

Area of Science:

  • Computer Vision
  • Machine Learning
  • Computational Neuroscience

Background:

  • Human action recognition in videos is a key research area in computer vision.
  • Traditional methods involve complex preprocessing followed by simple classification.
  • Focusing on the classifier stage offers potential for improved efficiency.

Purpose of the Study:

  • To develop an efficient human action recognition method using reservoir computing.
  • To introduce a novel training method for reservoir computers that integrates short and long time scales.
  • To evaluate the proposed algorithm's performance on a standard dataset and explore its hardware implementation.

Main Methods:

  • Utilized reservoir computing, a machine learning algorithm.
  • Introduced a new training method termed "Timesteps Of Interest" (TOI).
  • Evaluated performance using numerical simulations and a photonic implementation on the KTH dataset.

Main Results:

  • Achieved high accuracy and speed in human action recognition.
  • Demonstrated the capability for real-time processing of multiple video streams.
  • Validated the effectiveness of the TOI training method for reservoir computers.

Conclusions:

  • The proposed reservoir computing approach offers an efficient solution for human action recognition.
  • The TOI training method effectively combines short and long time scales for improved performance.
  • This work is a significant step towards developing dedicated hardware for real-time video processing.