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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We present an innovative two-headed attention layer that combines geometricand latent features to segment a 3D scene into semantically meaningful subsets.Each head combines local and global information, using either the geometric orlatent features, of a neighborhood of points and uses this information to learnbetter local relationships. This Geometric-Latent attention layer (Ge-Latto) iscombined with a sub-sampling strategy to capture global features. Our method isinvariant to permutation thanks to the use of shared-MLP layers, and it canalso be used with point clouds with varying densities because the localattention layer does not depend on the neighbor order. Our proposal is simpleyet robust, which allows it to achieve competitive results in the ShapeNetPartand ModelNet40 datasets, and the state-of-the-art when segmenting the complexdataset S3DIS, with 69.2% IoU on Area 5, and 89.7% overall accuracy usingK-fold cross-validation on the 6 areas.

Two Heads are Better than One: Geometric-Latent Attention for Point Cloud Classification and Segmentation