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

Roof plane segmentation from airborne LiDAR point clouds is an importanttechnology for 3D building model reconstruction. One of the key issues of planesegmentation is how to design powerful features that can exactly distinguishadjacent planar patches. The quality of point feature directly determines theaccuracy of roof plane segmentation. Most of existing approaches usehandcrafted features to extract roof planes. However, the abilities of thesefeatures are relatively low, especially in boundary area. To solve thisproblem, we propose a boundary-aware point clustering approach in Euclidean andembedding spaces constructed by a multi-task deep network for roof planesegmentation. We design a three-branch network to predict semantic labels,point offsets and extract deep embedding features. In the first branch, weclassify the input data as non-roof, boundary and plane points. In the secondbranch, we predict point offsets for shifting each point toward its respectiveinstance center. In the third branch, we constrain that points of the sameplane instance should have the similar embeddings. We aim to ensure that pointsof the same plane instance are close as much as possible in both Euclidean andembedding spaces. However, although deep network has strong featurerepresentative ability, it is still hard to accurately distinguish points nearplane instance boundary. Therefore, we first group plane points into manyclusters in the two spaces, and then we assign the rest boundary points totheir closest clusters to generate final complete roof planes. In this way, wecan effectively reduce the influence of unreliable boundary points. Inaddition, we construct a synthetic dataset and a real dataset to train andevaluate our approach. The experiments results show that the proposed approachsignificantly outperforms the existing state-of-the-art approaches.

A boundary-aware point clustering approach in Euclidean and embedding spaces for roof plane segmentation