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 Videos

PRIN/SPRIN: On Extracting Point-Wise Rotation Invariant Features.

Yang You, Yujing Lou, Ruoxi Shi

    IEEE Transactions on Pattern Analysis and Machine Intelligence
    |November 25, 2021
    PubMed
    Summary

    This study introduces PRIN, a novel point-set learning framework for rotation invariant point cloud analysis. Its sparse version, SPRIN, achieves superior performance in tasks like classification and segmentation without data augmentation.

    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

    Semantic Frame Interpolation.

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

    Probing Effective and Efficient Category-Level Articulated Object Pose Perception.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same author

    EvolveNav: Empowering LLM-Based Vision-Language Navigation via Self-Improving Embodied Reasoning.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same author

    Integrated multi-omics analysis unveils microbiota-metabolite-host interactions and novel biomarkers for early diabetic kidney disease diagnosis.

    Frontiers in immunology·2026
    Same author

    From sparse semantics to rich instances: Empowering label-efficient LiDAR panoptic segmentation via geometric priors.

    Neural networks : the official journal of the International Neural Network Society·2026
    Same author

    Neural Eigenfunctions are Structured Representation Learners.

    IEEE transactions on pattern analysis and machine intelligence·2025

    Area of Science:

    • Computer Vision
    • Machine Learning
    • 3D Data Analysis

    Background:

    • Point cloud analysis is hindered by unknown orientations, posing challenges for real-world applications.
    • Existing methods often require pose priors or extensive data augmentation to handle orientation variations.

    Purpose of the Study:

    • To develop a novel point-set learning framework, PRIN (Point-wise Rotation Invariant Network), for robust rotation invariant feature extraction in point clouds.
    • To extend PRIN to a sparse version, SPRIN, for efficient processing of sparse point cloud data.

    Main Methods:

    • Constructing spherical signals using Density Aware Adaptive Sampling to address distorted point distributions.
    • Employing Spherical Voxel Convolution and Point Re-sampling for rotation invariant feature extraction at the point level.

    Related Experiment Videos

  • Developing SPRIN for direct operation on sparse point clouds, enhancing computational efficiency.

    • Main Results:

    • SPRIN demonstrates superior performance on randomly rotated point cloud datasets compared to state-of-the-art methods, without requiring data augmentation.
    • The proposed methods achieve point-wise rotation invariance, validated through theoretical proofs and analysis.
    • PRIN and SPRIN show versatility, applicable to object classification, part segmentation, 3D feature matching, and label alignment.

    Conclusions:

    • PRIN and SPRIN offer a powerful solution for rotation invariant point cloud analysis, overcoming limitations of existing methods.
    • The framework's ability to handle arbitrary orientations without prior information significantly advances the field.
    • Publicly available code will facilitate reproducibility and further research in 3D point cloud understanding.