Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jun 3, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Group-aware and position-interactive learning for point cloud robust segmentation in subterranean environments.

Mengting Liu¹, Haijiang Zhu¹, Jian Cheng^2,3

¹College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, 100029, China.

Scientific Reports

|June 1, 2026

Summary

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

Morphological map of the proximal ulna bare area: a computer-assisted anatomical study in relation to olecranon osteotomy.

Journal of shoulder and elbow surgery·2025

Same author

The Path Planning and Location Method of Inspection Robot in a Large Storage Tank Bottom.

Computational intelligence and neuroscience·2023

Same author

Combined CNN and Pixel Feature Image for Fatty Liver Ultrasound Image Classification.

Computational and mathematical methods in medicine·2022

Same author

Oxygen vacancies engineering in electrocatalysts nitrogen reduction reaction.

Frontiers in chemistry·2022

Same author

Monocular Depth Estimation: Lightweight Convolutional and Matrix Capsule Feature-Fusion Network.

Sensors (Basel, Switzerland)·2022

Same author

Transfer RNA-derived small RNA: A rising star in oncology.

Seminars in cancer biology·2021

Same journal

Correction: A method for supervoxel-wise association studies of age and other non-imaging variables from coronary computed tomography angiograms.

Scientific reports·2026

Same journal

Poly(bromophenol blue)/CoSn(OH)<sub>6</sub> cubic particles modified pencil graphite electrode for electrochemical determination of diphenhydramine.

Scientific reports·2026

Same journal

Dietary Chlorella, Spirulina, and acidifier modulate jejunal cytokine-related gene expression in broiler chickens.

Scientific reports·2026

Same journal

Perceived physical activity barriers in university students: associations with fatigue and eating behaviours.

Scientific reports·2026

Same journal

Refuge limitation structures habitat use in agricultural landscapes: evidence from Sunda pangolins.

Scientific reports·2026

Same journal

Lightweight stateless transaction verification with outsourced witness updates for UTXO blockchains.

Scientific reports·2026

See all related articles

This summary is machine-generated.

This study introduces an efficient point cloud semantic segmentation framework for underground environments. The method enhances 3D perception in challenging conditions like coal mines, improving safety monitoring and infrastructure inspection.

Area of Science:

Computer Vision
Geospatial AI
Robotics

Background:

Underground infrastructure demands high-precision 3D perception for safety and inspection.
Existing point cloud segmentation methods struggle with underground environments due to noise, blurred boundaries, and repetitive patterns.
Computational costs of current methods limit practical deployment in real-world underground scenarios.

Purpose of the Study:

To develop an efficient and robust point cloud semantic segmentation framework for complex underground infrastructure.
To address the limitations of generalization robustness and computational overhead in existing methods.
To enhance 3D perception capabilities for applications like disaster prevention and infrastructure inspection.

Main Methods:

Proposed a lightweight Group-aware Channel Interaction Module (GCIM) utilizing weight-shared grouped channel attention and inter-group communication.

Keywords:

Interactive positional encoding Lightweight group-aware channel interaction Point cloud segmentation

More Related Videos

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

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

Published on: November 30, 2022

Related Experiment Videos

Last Updated: Jun 3, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

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

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

Published on: November 30, 2022

Introduced Interactive Position Encoding (IPE) to dynamically integrate spatial context with point features using relative positional information.

Strengthened spatial-aware feature interaction to mitigate boundary ambiguity caused by noise and geometric similarity.

Main Results:

The proposed framework achieved superior performance on underground datasets (Coal Mines, Seg2Tunnel, OpenTrench3D) and general indoor datasets (ScanNetv2, S3DIS).
Demonstrated significant improvements in mean Intersection over Union (mIoU) and Overall Accuracy (OA).
Exhibited outstanding segmentation capability and cross-scene generalization ability.

Conclusions:

The developed framework offers an effective technical solution for intelligent 3D information extraction in underground environments.
The method provides crucial support for safety monitoring and infrastructure management in challenging subterranean settings.
The efficient and robust approach facilitates practical deployment for real-time underground perception tasks.