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

Association Areas of the Cortex01:21

Association Areas of the Cortex

8.8K
Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
8.8K
Collisions in Multiple Dimensions: Introduction01:05

Collisions in Multiple Dimensions: Introduction

6.5K
It is far more common for collisions to occur in two dimensions; that is, the initial velocity vectors are neither parallel nor antiparallel to each other. Let's see what complications arise from this. The first idea is that momentum is a vector. Like all vectors, it can be expressed as a sum of perpendicular components (usually, though not always, an x-component and a y-component, and a z-component if necessary). Thus, when the statement of conservation of momentum is written for a...
6.5K

You might also read

Related Articles

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

Sort by
Same author

Adaptive Step-Size Tuning Combined with Ternary Search Optimization for FCS-MPC Sensorless Speed Estimation of PMSMs.

Sensors (Basel, Switzerland)·2026
Same author

Near-infrared-driven self-propelled Janus nanomotors penetrate persistent biofilms to deliver anti-inflammatory and photothermal therapies for accelerated wound healing.

Biomaterials·2026
Same author

Dual-channel grayscale display and encryption by near-field interference of transmitted polarized light on a nanoprinting metasurface.

Optics express·2025
Same author

ELAVL1-stabilized USP22 promotes diabetic nephropathy progression via mediating podocyte injury and death by triggering ACSL4 deubiquitination.

Transplant immunology·2025
Same author

Nine high-quality Anas genomes provide insights into Anas evolution and domestication.

Cell reports·2025
Same author

Metabolic risks remain a serious threat to cardiovascular disease: findings from the Global Burden of Disease Study 2019.

Internal and emergency medicine·2024

Related Experiment Video

Updated: Jan 13, 2026

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

Multi-Modal Data Fusion for 3D Object Detection Using Dual-Attention Mechanism.

Mengying Han1,2, Benlan Shen1,2, Jiuhong Ruan1,2

  • 1School of Rail Transportation, Shandong Jiaotong University, Jinan 250357, China.

Sensors (Basel, Switzerland)
|October 29, 2025
PubMed
Summary

This study enhances 3D object detection for small objects using fused LiDAR and camera data with improved PointPillars. The new method significantly boosts detection accuracy for pedestrians and cyclists in autonomous driving.

Keywords:
3D object detectionmultimodal data fusionpillar-wise channel attentionspatial attention mechanism

More Related Videos

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

8.1K

Related Experiment Videos

Last Updated: Jan 13, 2026

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
A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

8.1K

Area of Science:

  • Computer Vision
  • Robotics
  • Artificial Intelligence

Background:

  • Small objects in 3D point clouds suffer from missing features due to sparsity and irregularity.
  • Weak feature representation in point clouds leads to poor detection performance for small objects.

Purpose of the Study:

  • To develop a multi-modal 3D object detection method to improve the detection of small objects.
  • To enhance feature representation by fusing LiDAR point clouds with camera images.

Main Methods:

  • Proposed an improved PointPillars framework integrating LiDAR and camera data.
  • Introduced a Pillar-wise Channel Attention (PCA) module to emphasize critical features.
  • Embedded a Spatial Attention Module (SAM) into the backbone network for enhanced spatial representation.

Main Results:

  • Significant improvements in small-object detection performance on the KITTI dataset compared to baseline PointPillars.
  • Average Precision (AP) for pedestrians and cyclists increased by 7.06% and 3.08% (BEV), and 4.36% and 2.58% (3D), respectively.
  • Outperformed existing methods in detecting small objects, with enhanced capabilities shown through visualizations.

Conclusions:

  • The multi-modal approach effectively enhances 3D object detection, especially for small objects.
  • The proposed method shows strong potential for complex autonomous driving scenarios.
  • Attention modules (PCA and SAM) are crucial for improving feature representation and detection accuracy.