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

Neural shape models can represent complex 3D shapes with a compact latentspace. When applied to dynamically deforming shapes such as the human hands,however, they would need to preserve temporal coherence of the deformation aswell as the intrinsic identity of the subject. These properties are difficultto regularize with manually designed loss functions. In this paper, we learn aneural deformation model that disentangles the identity-induced shapevariations from pose-dependent deformations using implicit neural functions. Weperform template-free unsupervised learning on 3D scans without explicit meshcorrespondence or semantic correspondences of shapes across subjects. We canthen apply the learned model to reconstruct partial dynamic 4D scans of novelsubjects performing unseen actions. We propose two methods to integrate globalpose alignment with our neural deformation model. Experiments demonstrate theefficacy of our method in the disentanglement of identities and pose. Ourmethod also outperforms traditional skeleton-driven models in reconstructingsurface details such as palm prints or tendons without limitations from a fixedtemplate.

Identity-Disentangled Neural Deformation Model for Dynamic Meshes