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/

Zendy Plus

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

Zendy Tools

Zendy Open

Human hand actions are quite complex, especially when they involve objectmanipulation, mainly due to the high dimensionality of the hand and the vastaction space that entails. Imitating those actions with dexterous hand modelsinvolves different important and challenging steps: acquiring human handinformation, retargeting it to a hand model, and learning a policy fromacquired data. In this work, we capture the hand information by using astate-of-the-art hand pose estimator. We tackle the retargeting problem fromthe hand pose to a 29 DoF hand model by combining inverse kinematics and PSOwith a task objective optimisation. This objective encourages the virtual handto accomplish the manipulation task, relieving the effect of the estimator'snoise and the domain gap. Our approach leads to a better success rate in thegrasping task compared to our inverse kinematics baseline, allowing us torecord successful human demonstrations. Furthermore, we used thesedemonstrations to learn a policy network using generative adversarial imitationlearning (GAIL) that is able to autonomously grasp an object in the virtualspace.

Task-Oriented Hand Motion Retargeting for Dexterous Manipulation Imitation