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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Due to the costliness of labelled data in real-world applications,semi-supervised object detectors, underpinned by pseudo labelling, areappealing. However, handling confusing samples is nontrivial: discardingvaluable confusing samples would compromise the model generalisation whileusing them for training would exacerbate the confirmation bias issue caused byinevitable mislabelling. To solve this problem, this paper proposes to useconfusing samples proactively without label correction. Specifically, a virtualcategory (VC) is assigned to each confusing sample such that they can safelycontribute to the model optimisation even without a concrete label. It isattributed to specifying the embedding distance between the training sample andthe virtual category as the lower bound of the inter-class distance. Moreover,we also modify the localisation loss to allow high-quality boundaries forlocation regression. Extensive experiments demonstrate that the proposed VClearning significantly surpasses the state-of-the-art, especially with smallamounts of available labels.

Semi-supervised Object Detection via Virtual Category Learning