Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

Neural Blend-Field

A Neural Blend-Field is an implicit neural representation architecture used in 3D shape and deformation modeling that adaptively blends multiple localized coordinate-based neural fields to represent complex, continuous geometries. Instead of relying on a single monolithic network to model an entire 3D surface or motion field, this method decomposes a shape into semantically meaningful local regions governed by dedicated sub-networks. A lightweight neural fusion module then dynamically weights and combines the outputs of these local implicit functions based on spatial query coordinates. By distributing geometric or kinematic features across coordinated local fields and learning smooth transitions between them, a Neural Blend-Field significantly improves the ability of implicit neural models to capture fine-grained, high-frequency details while mitigating computational complexity and parameter overhead.

1 item

ImFace: A Nonlinear 3D Morphable Face Model with Implicit Neural Representations

ImFace: A Nonlinear 3D Morphable Face Model with Implicit Neural Representations

Mingwu Zheng, Hongyu Yang, Di Huang, Liming Chen

OrganizationsBeihang UniversityÉcole Centrale de LyonLIRIS

Why you should read this

Proposes a nonlinear 3D morphable face model based on implicit neural representations that explicitly disentangles identity and expression deformation fields, enabling high-fidelity face reconstruction and synthesis directly from non-watertight surfaces.

Precise representations of 3D faces are beneficial to various computer vision and graphics applications. Due to the data discretization and model linearity, however, it remains challenging to capture accurate identity and expression clues in current studies. This paper presents a novel 3D morphable face model, namely ImFace, to learn a nonlinear and continuous space with implicit neural representations. It builds two explicitly disentangled deformation fields to model complex shapes associated with identities and expressions, respectively, and designs an improved learning strategy to extend embeddings of expressions to allow more diverse changes. We further introduce a Neural Blend-Field to learn sophisticated details by adaptively blending a series of local fields. In addition to ImFace, an effective preprocessing pipeline is proposed to address the issue of watertight input requirement in implicit representations, enabling them to work with common facial surfaces for the first time. Extensive experiments are performed to demonstrate the superiority of ImFace.

Added

2026-09-26