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 Experiment Video

Updated: Nov 24, 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

815

Cascaded Cross-Modality Fusion Network for 3D Object Detection.

Zhiyu Chen1, Qiong Lin2, Jing Sun1

  • 1School of Computer Science, Nanjing University of Posts and Telecommunications, No. 9 Wenyuan Road, Yadong New District, Nanjing 210023, China.

Sensors (Basel, Switzerland)
|December 22, 2020
PubMed
Summary
This summary is machine-generated.

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

A CTC-Based Speech Recognition Network Fusing Local Convolution and Global Attention.

Sensors (Basel, Switzerland)·2026
Same author

Computer Vision-Guided Automated External Defibrillator System Improves Cardiopulmonary Resuscitation Accuracy: A Randomized Crossover Trial.

Health science reports·2026
Same author

A Progressive Feature Learning Network for <i>Cordyceps sinensis</i> Image Recognition.

Sensors (Basel, Switzerland)·2025
Same author

A Deep-Learning-Based CPR Action Standardization Method.

Sensors (Basel, Switzerland)·2024
Same author

ISLS: An Illumination-Aware Sauce-Packet Leakage Segmentation Method.

Sensors (Basel, Switzerland)·2024
Same author

NRVC: Neural Representation for Video Compression with Implicit Multiscale Fusion Network.

Entropy (Basel, Switzerland)·2023

This paper introduces CCFNet for LIDAR-RGB fusion 3D object detection, improving accuracy by addressing data misalignment and optimizing bounding box regression. The novel approach enhances object representation and performance on benchmarks.

Area of Science:

  • Computer Vision
  • Robotics
  • Sensor Fusion

Background:

  • 3D object detection using LIDAR-RGB fusion faces challenges due to data format differences and sensor misalignment.
  • Traditional Intersection over Union (IoU) optimization in bounding box regression can lead to biased back-propagation, especially in non-overlapping cases.

Purpose of the Study:

  • To propose a novel network, the Cascaded Cross-Modality Fusion Network (CCFNet), to address the challenges in LIDAR-RGB fusion-based 3D object detection.
  • To enhance the discriminative representation of objects by effectively fusing geometric and semantic features from LIDAR and RGB data, respectively.

Main Methods:

  • Developed a Cascaded Multi-scale Fusion (CMF) module to integrate multi-scale point cloud features into RGB streams, reinforcing object representation.
Keywords:
3D object detectionLIDARmulti-sensor fusionpoint cloud processing

More Related Videos

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.5K

Related Experiment Videos

Last Updated: Nov 24, 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

815
Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.5K
  • Introduced a novel Center 3D IoU loss function that incorporates anchor center distances to improve bounding box regression, particularly for non-overlapping scenarios.

    • Main Results:

    • The proposed CCFNet demonstrated superior performance compared to existing methods on the KITTI benchmark.
    • The CMF module effectively fuses LIDAR geometric and RGB semantic information, leading to improved object detection.

    Conclusions:

    • CCFNet offers a robust solution for LIDAR-RGB fusion 3D object detection by effectively handling data misalignment and optimizing regression losses.
    • The proposed fusion strategy and loss function significantly advance the state-of-the-art in autonomous driving perception systems.