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

You might also read

Related Articles

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

Sort by
Same author

Improved CenterNet-Based Multimodal Object Detection for Low-Light and Complex Environments.

Sensors (Basel, Switzerland)·2026
Same author

Research on Q-Learning-Based Cooperative Optimization Methodology for Dynamic Task Scheduling and Energy Consumption in Underwater Pan-Tilt Systems.

Sensors (Basel, Switzerland)·2025
Same author

Developmental Stage-Specific Embryonic Induction of HepG2 Cell Differentiation.

Digestive diseases and sciences·2015
Same author

Long-Term Effects of Pericardiectomy on Left Ventricular Mechanics Evaluated by Using Speckle Tracking Echocardiography in Patients with Constrictive Pericarditis.

Ultrasound in medicine & biology·2015
Same author

NMR-Based Metabolomics Reveal a Recovery from Metabolic Changes in the Striatum of 6-OHDA-Induced Rats Treated with Basic Fibroblast Growth Factor.

Molecular neurobiology·2015
Same author

Carbonic anhydrase mimics for enhanced CO2 absorption in an amine-based capture solvent.

Dalton transactions (Cambridge, England : 2003)·2015

Related Experiment Video

Updated: May 25, 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

452

SPS-RCNN: Semantic-Guided Proposal Sampling for 3D Object Detection from LiDAR Point Clouds.

Hengxin Xu1, Lei Yang2,3, Shengya Zhao3

  • 1College of Transportation, Shandong University of Science and Technology, Qingdao 266590, China.

Sensors (Basel, Switzerland)
|February 26, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces Semantic-Guided Proposal Sampling-RCNN (SPS-RCNN), a novel framework for 3D object detection using LiDAR. SPS-RCNN enhances foreground point detection and improves accuracy for distant objects.

Keywords:
3D object detectioncascade networklight detection and ranging (lidar)point–voxel fusionsemantic-guided proposal sampling

More Related Videos

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

348

Related Experiment Videos

Last Updated: May 25, 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

452
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

348

Area of Science:

  • Computer Vision
  • LiDAR 3D Object Detection
  • Machine Learning

Background:

  • LiDAR-based 3D object detection is crucial for autonomous systems due to its robustness in varying lighting and detailed geometric capture.
  • Existing methods struggle with a high ratio of background points and reduced accuracy for distant object detection.

Purpose of the Study:

  • To develop an improved 3D object detection framework addressing the limitations of current LiDAR-based methods.
  • To enhance the accuracy and robustness of detecting objects at various ranges, particularly distant ones.

Main Methods:

  • Proposed Semantic-Guided Proposal Sampling-RCNN (SPS-RCNN), a multi-stage point-voxel fusion detection framework.
  • Introduced a novel Semantic-Guided Proposal Sampling (SPS) method within the Keypoint Sampling Stream (KSS) for increased foreground point ratio and outlier sensitivity.
  • Utilized a Cascade Attention Module (CAM) in the Progressive Refinement Network (PRN) for aggregating multi-subnet features and refining proposals.

Main Results:

  • SPS-RCNN demonstrated improved detection accuracy on the KITTI dataset.
  • The framework exhibited enhanced robustness across different object categories compared to baseline methods.
  • The SPS method effectively increased foreground point representation and improved detection of medium- and long-range objects.

Conclusions:

  • SPS-RCNN offers a significant advancement in LiDAR-based 3D object detection.
  • The proposed semantic-guided sampling and progressive refinement strategies enhance detection performance and robustness.