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/

Zendy Plus

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

Zendy Tools

Zendy Open

Deep learning based methods have significantly boosted the study of automaticbuilding extraction from remote sensing images. However, delineating vectorizedand regular building contours like a human does remains very challenging, dueto the difficulty of the methodology, the diversity of building structures, andthe imperfect imaging conditions. In this paper, we propose the firstend-to-end learnable building contour extraction framework, named BuildMapper,which can directly and efficiently delineate building polygons just as a humandoes. BuildMapper consists of two main components: 1) a contour initializationmodule that generates initial building contours; and 2) a contour evolutionmodule that performs both contour vertex deformation and reduction, whichremoves the need for complex empirical post-processing used in existingmethods. In both components, we provide new ideas, including a learnablecontour initialization method to replace the empirical methods, dynamicpredicted and ground truth vertex pairing for the static vertex correspondenceproblem, and a lightweight encoder for vertex information extraction andaggregation, which benefit a general contour-based method; and a well-designedvertex classification head for building corner vertices detection, which castslight on direct structured building contour extraction. We also built asuitable large-scale building dataset, the WHU-Mix (vector) building dataset,to benefit the study of contour-based building extraction methods. Theextensive experiments conducted on the WHU-Mix (vector) dataset, the WHUdataset, and the CrowdAI dataset verified that BuildMapper can achieve astate-of-the-art performance, with a higher mask average precision (AP) andboundary AP than both segmentation-based and contour-based methods.

BuildMapper: A Fully Learnable Framework for Vectorized Building Contour Extraction