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

keyword

deformation learning

Deformation learning is a computational approach in computer vision and geometric deep learning that trains machine learning models to predict and represent continuous non-rigid transformations of shapes, surfaces, or spatial fields. Rather than treating geometric variations as static structures or relying on rigid and linear approximations, deformation learning uses neural networks to parameterize coordinate displacement fields that map a reference or canonical shape space to various deformed target states. This framework allows models to disentangle and control distinct modes of variation, such as identity-specific traits and dynamic expressions or poses, facilitating tasks such as non-rigid shape registration, 3D morphable modeling, animation, and generative reconstruction while maintaining spatial and topological consistency.

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