English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Point cloud analysis is challenging due to the irregularity of the pointcloud data structure. Existing works typically employ the ad-hocsampling-grouping operation of PointNet++, followed by sophisticated localand/or global feature extractors for leveraging the 3D geometry of the pointcloud. Unfortunately, the sampling-grouping operations do not address the pointcloud's irregularity, whereas the intricate local and/or global featureextractors led to poor computational efficiency. In this paper, we introduce anovel DualNorm module after the sampling-grouping operation to effectively andefficiently address the irregularity issue. The DualNorm module consists ofPoint Normalization, which normalizes the grouped points to the sampled points,and Reverse Point Normalization, which normalizes the sampled points to thegrouped points. The proposed framework, PointNorm, utilizes local mean andglobal standard deviation to benefit from both local and global features whilemaintaining a faithful inference speed. Experiments show that we achievedexcellent accuracy and efficiency on ModelNet40 classification, ScanObjectNNclassification, ShapeNetPart Part Segmentation, and S3DIS SemanticSegmentation. Code is available athttps://github.com/ShenZheng2000/PointNorm-for-Point-Cloud-Analysis.

PointNorm: Dual Normalization is All You Need for Point Cloud Analysis