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

Parametric 3D body models like SMPL only represent minimally-clothed peopleand are hard to extend to clothing because they have a fixed mesh topology andresolution. To address these limitations, recent work uses implicit surfaces orpoint clouds to model clothed bodies. While not limited by topology, suchmethods still struggle to model clothing that deviates significantly from thebody, such as skirts and dresses. This is because they rely on the body tocanonicalize the clothed surface by reposing it to a reference shape.Unfortunately, this process is poorly defined when clothing is far from thebody. Additionally, they use linear blend skinning to pose the body and theskinning weights are tied to the underlying body parts. In contrast, we modelthe clothing deformation in a local coordinate space without canonicalization.We also relax the skinning weights to let multiple body parts influence thesurface. Specifically, we extend point-based methods with a coarse stage, thatreplaces canonicalization with a learned pose-independent "coarse shape" thatcan capture the rough surface geometry of clothing like skirts. We then refinethis using a network that infers the linear blend skinning weights and posedependent displacements from the coarse representation. The approach works wellfor garments that both conform to, and deviate from, the body. We demonstratethe usefulness of our approach by learning person-specific avatars fromexamples and then show how they can be animated in new poses and motions. Wealso show that the method can learn directly from raw scans with missing data,greatly simplifying the process of creating realistic avatars. Code isavailable for research purposes at{\small\url{https://qianlim.github.io/SkiRT}}.

Neural Point-based Shape Modeling of Humans in Challenging Clothing