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

Machine learning models of visual action recognition are typically trainedand tested on data from specific situations where actions are associated withcertain objects. It is an open question how action-object associations in thetraining set influence a model's ability to generalize beyond trainedsituations. We set out to identify properties of training data that lead toaction recognition models with greater generalization ability. To do this, wetake inspiration from a cognitive mechanism called cross-situational learning,which states that human learners extract the meaning of concepts by observinginstances of the same concept across different situations. We performcontrolled experiments with various types of action-object associations, andidentify key properties of action-object co-occurrence in training data thatlead to better classifiers. Given that these properties are missing in thedatasets that are typically used to train action classifiers in the computervision literature, our work provides useful insights on how we should bestconstruct datasets for efficiently training for better generalization.

Action Recognition based on Cross-Situational Action-object Statistics