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: Aug 30, 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

616

3D Cascade RCNN: High Quality Object Detection in Point Clouds.

Qi Cai, Yingwei Pan, Ting Yao

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |August 30, 2022
    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

    MAFB Overexpression in Macrophages Promotes Wound Healing in Diabetic Foot Ulcer by Transcriptionally Activating WDR74 to Drive a Tissue-Repair Phenotype and Suppress Inflammation and Oxidative Stress.

    Applied biochemistry and biotechnology·2026
    Same author

    Kilogram-Scale Synthesis of Few-Layer Graphene-Skinned Copper Powders via Pulsed-Pressure Chemical Vapor Deposition.

    Small (Weinheim an der Bergstrasse, Germany)·2026
    Same author

    Genome-Wide association study of genetic variants influencing creatine kinase levels in Chinese winter sports athletes.

    Gene·2026
    Same author

    Vegetation succession enhances soil organic carbon sequestration by modulating microbial carbon cycling genes in coastal marshes, eastern China.

    Marine pollution bulletin·2026
    Same author

    Plaque phenotypes across remodeling patterns in vertebrobasilar atherosclerosis: a high-resolution magnetic resonance vessel wall imaging study.

    Quantitative imaging in medicine and surgery·2026
    Same author

    Surface Functionalization of 3D-Printed Porous Ti-6Al-4V Scaffolds with PDA‑mHA-Col I Composite Coating Drives Macrophage Polarization toward the M2 Phenotype and Promotes Immunomodulation: An In Vitro Study.

    Current gene therapy·2026
    Same journal

    Style-Aware Contrastive Test-Time Adaptation: A Dual-Cache Model for Robust Vision-Language Alignment.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    Same journal

    Semantic Frame Interpolation.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    Same journal

    Physics-Guided Cross-Modal Decoupling with Test-Time Adaptation for Hyperspectral Image Restoration.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    Same journal

    Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    Same journal

    AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    Same journal

    BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    See all related articles

    We introduce 3D Cascade RCNN, a novel cascade architecture for 3D object detection using sparse LiDAR data. It improves detection quality by progressively refining proposals and re-weighting based on point completeness, outperforming existing methods.

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Robotics

    Background:

    • 2D object detection has advanced with Cascade RCNN, utilizing sequential detectors for improved proposal quality.
    • 3D object detection with sparse LiDAR data presents unique challenges, lacking established cascade structures.

    Purpose of the Study:

    • To develop a novel cascade architecture, 3D Cascade RCNN, for high-quality 3D object detection.
    • To address the challenges posed by sparse LiDAR point clouds in 3D detection.

    Main Methods:

    • Proposed a cascade architecture allocating multiple detectors to voxelized point clouds.
    • Introduced 'point completeness score' to quantify sparsity within 3D bounding boxes.
    • Implemented completeness-aware re-weighting to guide stage detector learning, prioritizing proposals with denser point distributions.

    More Related Videos

    Deep Neural Networks for Image-Based Dietary Assessment
    13:19

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    9.3K
    Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
    10:23

    Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans

    Published on: September 8, 2023

    3.0K

    Related Experiment Videos

    Last Updated: Aug 30, 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

    616
    Deep Neural Networks for Image-Based Dietary Assessment
    13:19

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    9.3K
    Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
    10:23

    Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans

    Published on: September 8, 2023

    3.0K

    Main Results:

    • 3D Cascade RCNN demonstrates superior performance compared to state-of-the-art 3D object detection techniques.
    • The completeness-aware re-weighting effectively handles sparse input data without increasing computational cost (FLOPs).
    • Validation conducted on both KITTI and Waymo Open Datasets confirmed the effectiveness of the proposed method.

    Conclusions:

    • 3D Cascade RCNN offers an effective and simple cascade approach for 3D object detection in sparse LiDAR environments.
    • The proposed point completeness-aware re-weighting mechanism is crucial for adapting cascade paradigms to sparse data challenges.
    • The method achieves state-of-the-art results, providing a valuable contribution to 3D perception research.