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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Implicit functions represented as deep learning approximations are powerfulfor reconstructing 3D surfaces. However, they can only produce static surfacesthat are not controllable, which provides limited ability to modify theresulting model by editing its pose or shape parameters. Nevertheless, suchfeatures are essential in building flexible models for both computer graphicsand computer vision. In this work, we present methodology that combinesdetail-rich implicit functions and parametric representations in order toreconstruct 3D models of people that remain controllable and accurate even inthe presence of clothing. Given sparse 3D point clouds sampled on the surfaceof a dressed person, we use an Implicit Part Network (IP-Net)to jointly predictthe outer 3D surface of the dressed person, the and inner body surface, and thesemantic correspondences to a parametric body model. We subsequently usecorrespondences to fit the body model to our inner surface and then non-rigidlydeform it (under a parametric body + displacement model) to the outer surfacein order to capture garment, face and hair detail. In quantitative andqualitative experiments with both full body data and hand scans we show thatthe proposed methodology generalizes, and is effective even given incompletepoint clouds collected from single-view depth images. Our models and code canbe downloaded from http://virtualhumans.mpi-inf.mpg.de/ipnet.

Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction