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

This paper proposes a novel method for understanding daily hand-objectmanipulation by developing computer vision-based techniques. Specifically, wefocus on recognizing hand grasp types, object attributes and manipulationactions within an unified framework by exploring their contextualrelationships. Our hypothesis is that it is necessary to jointly model hands,objects and actions in order to accurately recognize multiple tasks that arecorrelated to each other in hand-object manipulation. In the proposed model, weexplore various semantic relationships between actions, grasp types and objectattributes, and show how the context can be used to boost the recognition ofeach component. We also explore the spatial relationship between the hand andobject in order to detect the manipulated object from hand in clutteredenvironment. Experiment results on all three recognition tasks show that ourproposed method outperforms traditional appearance-based methods which are notdesigned to take into account contextual relationships involved in hand-objectmanipulation. The visualization and generalizability study of the learnedcontext further supports our hypothesis.

Understanding hand-object manipulation by modeling the contextual relationship between actions, grasp types and object attributes