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

Many modern applications use computer vision to detect and count objects inmassive image collections. However, when the detection task is very difficultor in the presence of domain shifts, the counts may be inaccurate even withsignificant investments in training data and model development. We proposeDISCount -- a detector-based importance sampling framework for counting inlarge image collections that integrates an imperfect detector withhuman-in-the-loop screening to produce unbiased estimates of counts. We proposetechniques for solving counting problems over multiple spatial or temporalregions using a small number of screened samples and estimate confidenceintervals. This enables end-users to stop screening when estimates aresufficiently accurate, which is often the goal in a scientific study. On thetechnical side we develop variance reduction techniques based on controlvariates and prove the (conditional) unbiasedness of the estimators. DISCountleads to a 9-12x reduction in the labeling costs over naive screening for taskswe consider, such as counting birds in radar imagery or estimating damagedbuildings in satellite imagery, and also surpasses alternative covariate-basedscreening approaches in efficiency.

DISCount: Counting in Large Image Collections with Detector-Based Importance Sampling