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

Innovations in computer vision algorithms for satellite image analysis canenable us to explore global challenges such as urbanization and land use changeat the planetary level. However, domain shift problems are a common occurrencewhen trying to replicate models that drive these analyses to new areas,particularly in the developing world. If a model is trained with imagery andlabels from one location, then it usually will not generalize well to newlocations where the content of the imagery and data distributions aredifferent. In this work, we consider the setting in which we have a singlelarge satellite imagery scene over which we want to solve an applied problem --building footprint segmentation. Here, we do not necessarily need to worryabout creating a model that generalizes past the borders of our scene but caninstead train a local model. We show that surprisingly few labels are needed tosolve the building segmentation problem with very high-resolution (0.5m/px)satellite imagery with this setting in mind. Our best model trained with just527 sparse polygon annotations (an equivalent of 1500 x 1500 densely labeledpixels) has a recall of 0.87 over held out footprints and a R2 of 0.93 on thetask of counting the number of buildings in 200 x 200-meter windows. We applyour models over high-resolution imagery in Amman, Jordan in a case study onurban change detection.

Fast building segmentation from satellite imagery and few local labels