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

In this paper, we introduce a new building dataset and propose a novel domaingeneralization method to facilitate the development of building extraction fromhigh-resolution remote sensing images. The problem with the current buildingdatasets involves that they lack diversity, the quality of the labels isunsatisfactory, and they are hardly used to train a building extraction modelwith good generalization ability, so as to properly evaluate the realperformance of a model in practical scenes. To address these issues, we built adiverse, large-scale, and high-quality building dataset named the WHU-Mixbuilding dataset, which is more practice-oriented. The WHU-Mix building datasetconsists of a training/validation set containing 43,727 diverse imagescollected from all over the world, and a test set containing 8402 images fromfive other cities on five continents. In addition, to further improve thegeneralization ability of a building extraction model, we propose a domaingeneralization method named batch style mixing (BSM), which can be embedded asan efficient plug-and-play module in the frond-end of a building extractionmodel, providing the model with a progressively larger data distribution tolearn data-invariant knowledge. The experiments conducted in this studyconfirmed the potential of the WHU-Mix building dataset to improve theperformance of a building extraction model, resulting in a 6-36% improvement inmIoU, compared to the other existing datasets. The adverse impact of theinaccurate labels in the other datasets can cause about 20% IoU decrease. Theexperiments also confirmed the high performance of the proposed BSM module inenhancing the generalization ability and robustness of a model, exceeding thebaseline model without domain generalization by 13% and the recent domaingeneralization methods by 4-15% in mIoU.

A diverse large-scale building dataset and a novel plug-and-play domain generalization method for building extraction