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

Generation of Action Potential in Skeletal Muscles01:24

Generation of Action Potential in Skeletal Muscles

8.5K
Every cell in the body maintains a membrane potential due to an uneven distribution of positive and negative charges across its plasma membrane. The membrane potential is measured in millivolts and quantifies the difference in charge across the membrane.
Like neurons, muscle cells are also regarded as excitable due to their capacity to change in response to stimuli, primarily due to voltage-gated ion channels embedded in their plasma membranes, which get activated by alterations in the...
8.5K
Convolution Properties II01:17

Convolution Properties II

583
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
583
Convolution Properties I01:20

Convolution Properties I

571
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
571
Control Volume and System Representations01:16

Control Volume and System Representations

1.5K
Two key frameworks are employed to analyze mass, energy, and momentum transfer: the control volume approach and the system approach. These frameworks offer different perspectives, depending on whether the focus is on a specific region in space (control volume approach) or a defined mass of fluid (system approach).
The control volume approach considers a stationary region in space through which fluid flows. This region is bounded by a control surface.  For instance, in the case of water...
1.5K
Linear time-invariant Systems01:23

Linear time-invariant Systems

893
A system is linear if it displays the characteristics of homogeneity and additivity, together termed the superposition property. This principle is fundamental in all linear systems. Linear time-invariant (LTI) systems include systems with linear elements and constant parameters.
The input-output behavior of an LTI system can be fully defined by its response to an impulsive excitation at its input. Once this impulse response is known, the system's reaction to any other input can be...
893
State Space Representation01:27

State Space Representation

543
The frequency-domain technique, commonly used in analyzing and designing feedback control systems, is effective for linear, time-invariant systems. However, it falls short when dealing with nonlinear, time-varying, and multiple-input multiple-output systems. The time-domain or state-space approach addresses these limitations by utilizing state variables to construct simultaneous, first-order differential equations, known as state equations, for an nth-order system.
Consider an RLC circuit, a...
543

You might also read

Related Articles

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

Sort by
Same author

VinDr-Mammo: A large-scale benchmark dataset for computer-aided diagnosis in full-field digital mammography.

Scientific data·2023
Same author

3D Object Detection for Self-Driving Cars Using Video and LiDAR: An Ablation Study.

Sensors (Basel, Switzerland)·2023
Same author

Detecting COVID-19 from digitized ECG printouts using 1D convolutional neural networks.

PloS one·2022
Same author

Learning to diagnose common thorax diseases on chest radiographs from radiology reports in Vietnamese.

PloS one·2022
Same author

3D Object Detection with SLS-Fusion Network in Foggy Weather Conditions.

Sensors (Basel, Switzerland)·2021
Same author

Detecting, Tracking and Counting People Getting On/Off a Metropolitan Train Using a Standard Video Camera.

Sensors (Basel, Switzerland)·2020

Related Experiment Video

Updated: Jan 25, 2026

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

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.9K

Spatio⁻Temporal Image Representation of 3D Skeletal Movements for View-Invariant Action Recognition with Deep

Huy Hieu Pham1,2, Houssam Salmane3, Louahdi Khoudour4

  • 1Cerema, Project team STI, 1 avenue du Colonel Roche, F-31400 Toulouse, France. huy-hieu.pham@cerema.fr.

Sensors (Basel, Switzerland)
|April 27, 2019
PubMed
Summary

This study introduces a novel Skeleton Posture-Motion Feature (SPMF) representation and a deep learning framework for 3D human action recognition. The proposed method achieves state-of-the-art performance with low computational cost.

Keywords:
3D human action recognitionAHED-CNNsDenseNetEnhanced-SPMFSPMFskeleton-based representation

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

1.0K
Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy
12:09

Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy

Published on: August 5, 2014

18.5K

Related Experiment Videos

Last Updated: Jan 25, 2026

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

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.9K
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

1.0K
Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy
12:09

Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy

Published on: August 5, 2014

18.5K

Area of Science:

  • Computer Vision
  • Machine Learning
  • Human-Computer Interaction

Background:

  • 3D human action recognition from skeleton sequences is challenging due to the need for robust, viewpoint-invariant, and computationally efficient motion representations.
  • Representing spatio-temporal patterns and learning discriminative features are key hurdles in this domain.

Purpose of the Study:

  • To propose a novel skeleton-based representation and a deep learning framework for improved 3D human action recognition.
  • To address the challenges of spatio-temporal pattern representation and feature learning for skeletal data.

Main Methods:

  • A new action map, Skeleton Posture-Motion Feature (SPMF), is proposed as a compact image representation of skeleton poses and motions.
  • Adaptive Histogram Equalization (AHE) enhances SPMF to create Enhanced-SPMF for better pattern recognition.
  • Deep Convolutional Neural Networks (DenseNet architecture) are employed for end-to-end learning and classification using Enhanced-SPMFs.

Main Results:

  • The proposed method, utilizing Enhanced-SPMFs and DenseNet, achieves state-of-the-art performance across four diverse benchmark datasets.
  • Demonstrated superior accuracy in recognizing individual actions, interactions, and handling multiview and large-scale data.
  • The approach requires low computational time for both training and inference.

Conclusions:

  • The novel SPMF representation combined with deep learning offers a highly effective and efficient solution for 3D human action recognition.
  • The Enhanced-SPMF approach significantly improves upon existing methods, providing robustness and accuracy.
  • This framework presents a promising direction for real-world applications requiring real-time skeleton-based action recognition.