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 technique for efficiently modeling dynamic humans byexplicifying the implicit neural fields via a Neural Explicit Surface (NES).Implicit neural fields have advantages over traditional explicitrepresentations in modeling dynamic 3D content from sparse observations andeffectively representing complex geometries and appearances. Implicit neuralfields defined in 3D space, however, are expensive to render due to the needfor dense sampling during volumetric rendering. Moreover, their memoryefficiency can be further optimized when modeling sparse 3D space. To overcomethese issues, the paper proposes utilizing Neural Explicit Surface (NES) toexplicitly represent implicit neural fields, facilitating memory andcomputational efficiency. To achieve this, the paper creates a fullydifferentiable conversion between the implicit neural fields and the explicitrendering interface of NES, leveraging the strengths of both implicit andexplicit approaches. This conversion enables effective training of the hybridrepresentation using implicit methods and efficient rendering by integratingthe explicit rendering interface with a newly proposed rasterization-basedneural renderer that only incurs a texture color query once for the initial rayinteraction with the explicit surface, resulting in improved inferenceefficiency. NES describes dynamic human geometries with pose-dependent neuralimplicit surface deformation fields and their dynamic neural textures both in2D space, which is a more memory-efficient alternative to traditional 3Dmethods, reducing redundancy and computational load. The comprehensiveexperiments show that NES performs similarly to previous 3D approaches, withgreatly improved rendering speed and reduced memory cost.

Explicifying Neural Implicit Fields for Efficient Dynamic Human Avatar Modeling via a Neural Explicit Surface