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

We study the problem of semantic segmentation of large-scale 3D point clouds.In recent years, significant research efforts have been directed toward localfeature aggregation, improved loss functions and sampling strategies. While thefundamental framework of point cloud semantic segmentation has been largelyoverlooked, with most existing approaches rely on the U-Net architecture bydefault. In this paper, we propose U-Next, a small but mighty frameworkdesigned for point cloud semantic segmentation. The key to this framework is tolearn multi-scale hierarchical representations from semantically similarfeature maps. Specifically, we build our U-Next by stacking multiple U-Net$L^1$ codecs in a nested and densely arranged manner to minimize the semanticgap, while simultaneously fusing the feature maps across scales to effectivelyrecover the fine-grained details. We also devised a multi-level deepsupervision mechanism to further smooth gradient propagation and facilitatenetwork optimization. Extensive experiments conducted on three large-scalebenchmarks including S3DIS, Toronto3D, and SensatUrban demonstrate thesuperiority and the effectiveness of the proposed U-Next architecture. OurU-Next architecture shows consistent and visible performance improvementsacross different tasks and baseline models, indicating its great potential toserve as a general framework for future research.

Small but Mighty: Enhancing 3D Point Clouds Semantic Segmentation with U-Next Framework