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

Depth Perception and Spatial Vision01:15

Depth Perception and Spatial Vision

673
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
673

You might also read

Related Articles

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

Sort by
Same author

Low-Quality Sensor Data-Based Semi-Supervised Learning for Medical Image Segmentation.

Sensors (Basel, Switzerland)·2024
Same author

A Multi-Objectives Genetic Algorithm Based Predictive Model and Strategy Optimization during SLM Process.

Materials (Basel, Switzerland)·2022
Same author

Real-Time and High-Fidelity Tracking of Lysosomal Dynamics with a Dicyanoisophorone-Based Fluorescent Probe.

Analytical chemistry·2021
Same author

Real-time tracking lysosomal pH changes under heatstroke and redox stress with a novel near-infrared emissive probe.

Talanta·2021
Same author

<i>In Vivo</i> Imaging and Tracking Carbon Monoxide-Releasing Molecule-3 with an NIR Fluorescent Probe.

ACS sensors·2021
Same author

A novel reaction-based fluorescence probe for rapid imaging of HClO in live cells, animals, and injured liver tissues.

Talanta·2020

Related Experiment Video

Updated: Jul 9, 2025

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

559

DTV-CNN: Neural network based on depth and thickness views for efficient 3D shape classification.

Qingfeng Xia1

  • 1Culham Centre for Fusion Energy, United Kingdom Atomic Energy Authority, OX14 3DB, United Kingdom.

Heliyon
|November 29, 2023
PubMed
Summary

This study introduces an efficient 2.5D convolutional neural network (CNN) for deep learning on 3D shapes. The method significantly reduces training time and computational cost for mechanical and electronic engineering design.

Keywords:
3D shape classificationAICADCNNDepth mapEDAIntelligent engineering designNeural networkThickness

More Related Videos

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

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.2K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Related Experiment Videos

Last Updated: Jul 9, 2025

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

559
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.2K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Area of Science:

  • Computer Science
  • Engineering
  • Artificial Intelligence

Background:

  • Deep learning on 3D shapes is crucial for advancing engineering design.
  • Existing methods often require significant computational resources and time.
  • Bridging the gap between 2D image deep learning and 3D shape analysis is needed.

Purpose of the Study:

  • To develop a fast and effective algorithm for deep learning on 3D shapes.
  • To propose a shallow 2.5D convolutional neural network (CNN) architecture.
  • To enable efficient 3D shape analysis for engineering applications.

Main Methods:

  • A 3D shape to 2D image projection algorithm was developed.
  • 3D geometry was compressed into 2D 'thickness' and 'depth' views.
  • A dual-channel grayscale image (DTV) fused these views for feature extraction.
  • A mixed CNN and multiple linear parameter (MLP) model was employed.

Main Results:

  • Achieved 92% validation accuracy on the ModelNet10 dataset.
  • Reduced training time by an order of magnitude compared to multi-view CNNs.
  • Obtained 97% accuracy on FreeCAD mechanical parts and 95% on KiCAD electronic parts.
  • Demonstrated training in tens of minutes on a laptop CPU.

Conclusions:

  • The proposed 2.5D CNN approach offers an efficient solution for 3D shape analysis.
  • This method significantly reduces computational cost and training time.
  • The approach is adaptable for various machine learning tasks in CAD geometry.