Jove
Visualize
Contact Us

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

Pose-Perceptive Convolution: Learning Geometry-Adaptive Receptive Fields for Robust 6D Pose Estimation.

Sensors (Basel, Switzerland)ยท2026
See all related articles
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 Experiment Video

Updated: Jan 18, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K

U-ResNet, a Novel Network Fusion Method for Image Classification and Segmentation.

Wenkai Li1, Zhe Gao1, Yaqing Song1

  • 1The College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China.

Sensors (Basel, Switzerland)
|September 13, 2025
PubMed
Summary
This summary is machine-generated.

We introduce U-ResNet, a novel parallel network structure that combines U-Net and ResNet for enhanced image classification and segmentation. This architecture achieves high accuracy and rapid convergence, outperforming current state-of-the-art models.

Keywords:
ResNetU-Netcomputer visionimage classificationimage segmentation

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

736
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

3.2K

Related Experiment Videos

Last Updated: Jan 18, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

736
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

3.2K

Area of Science:

  • Computer Vision
  • Deep Learning Architectures

Background:

  • Image classification and segmentation are key computer vision tasks, with ResNet and U-Net being prominent models for each, respectively.
  • Existing fusion methods often prioritize segmentation (U-Net), neglecting ResNet's classification strengths.

Purpose of the Study:

  • To propose a novel U-ResNet architecture integrating U-Net's UBlock and ResNet's ResBlock in parallel.
  • To enhance performance in both image classification and segmentation tasks.
  • To address the vanishing gradient problem and improve network convergence.

Main Methods:

  • A parallel U-ResNet structure combining UBlock (convolution-deconvolution) and ResBlock (residual) was developed.
  • The UBlock processes pixel-level features from varying resolutions.
  • The ResBlock incorporates Selected Upsampling (SU) for low-resolution data and an improved Efficient Upsampling Convolutional Block (EUCB*) with Channel Shuffle for convergence.

Main Results:

  • The U-ResNet architecture demonstrated rapid convergence and high accuracy on both classification and segmentation tasks.
  • It effectively mitigated the vanishing gradient problem.
  • Performance surpassed state-of-the-art (SOTA) models on diverse datasets.

Conclusions:

  • The proposed U-ResNet architecture offers a powerful solution for combined image classification and segmentation.
  • Its parallel design and novel modules show significant potential for advanced computer vision applications.
  • Ablation studies confirmed the efficacy of individual U-ResNet components.