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 an effort to reduce annotation costs in action recognition, unsupervisedvideo domain adaptation methods have been proposed that aim to adapt apredictive model from a labelled dataset (i.e., source domain) to an unlabelleddataset (i.e., target domain). In this work we address a more realisticscenario, called open-set video domain adaptation (OUVDA), where the targetdataset contains "unknown" semantic categories that are not shared with thesource. The challenge lies in aligning the shared classes of the two domainswhile separating the shared classes from the unknown ones. In this work wepropose to address OUVDA with an unified contrastive learning framework thatlearns discriminative and well-clustered features. We also propose avideo-oriented temporal contrastive loss that enables our method to bettercluster the feature space by exploiting the freely available temporalinformation in video data. We show that discriminative feature spacefacilitates better separation of the unknown classes, and thereby allows us touse a simple similarity based score to identify them. We conduct thoroughexperimental evaluation on multiple OUVDA benchmarks and show the effectivenessof our proposed method against the prior art.

Simplifying Open-Set Video Domain Adaptation with Contrastive Learning