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

The unprecedented availability of spatial and temporal high-resolutionsatellite image time series (SITS) for crop type mapping is believed tonecessitate deep learning architectures to accommodate challenges arising fromboth dimensions. Recent state-of-the-art deep learning models have shownpromising results by stacking spatial and temporal encoders. However, wepresent a Pyramid Time-Series Transformer (PTST) that operates solely on thetemporal dimension, i.e., neglecting the spatial dimension, can producesuperior results with a drastic reduction in GPU memory consumption and easyextensibility. Furthermore, we augment it to perform semi-supervised learningby proposing a classification-friendly VAE framework that introduces clusteringmechanisms into latent space and can promote linear separability therein.Consequently, a few principal axes of the latent space can explain the majorityof variance in raw data. Meanwhile, the VAE framework with proposed tweaks canmaintain competitive classification performance as its purely discriminativecounterpart when only $40\%$ of labelled data is used. We hope the proposedframework can serve as a baseline for crop classification with SITS for itsmodularity and simplicity.

Tampered VAE for Improved Satellite Image Time Series Classification