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

When watching a video, humans can naturally extract human actions from thesurrounding scene context, even when action-scene combinations are unusual.However, unlike humans, video action recognition models often learnscene-biased action representations from the spurious correlation in trainingdata, leading to poor performance in out-of-context scenarios. Whilescene-debiased models achieve improved performance in out-of-context scenarios,they often overlook valuable scene information in the data. Addressing thischallenge, we propose Disentangled VIdeo representations of Action and Scene(DEVIAS), which aims to achieve holistic video understanding. Disentangledaction and scene representations with our method could provide flexibility toadjust the emphasis on action or scene information depending on downstream taskand dataset characteristics. Disentangled action and scene representationscould be beneficial for both in-context and out-of-context video understanding.To this end, we employ slot attention to learn disentangled action and scenerepresentations with a single model, along with auxiliary tasks that furtherguide slot attention. We validate the proposed method on both in-contextdatasets: UCF-101 and Kinetics-400, and out-of-context datasets: SCUBA and HAT.Our proposed method shows favorable performance across different datasetscompared to the baselines, demonstrating its effectiveness in diverse videounderstanding scenarios.

DEVIAS: Learning Disentangled Video Representations of Action and Scene for Holistic Video Understanding