Skip to main content
SCULPTOR: Skeleton-Consistent Face Creation Using a Learned Parametric Generator
Model Parametric Skull Face Anatomically Consistent Shape Pose Expression Appearance Trait
[Madsen et al. 2018] ✅ ✅ ✅ ✅ ✅ ❌ ❌ ❌ ❌
[Gruber et al. 2020] ✅ ✅ ✅ ❌ ✅ ❌ ❌ ❌ ❌
[Ichim et al. 2017] ❌ ✅ ✅ ❌ ✅ ✅ ✅ ✅ ❌
[Li et al. 2020] ❌ ❌ ✅ ❌ ✅ ✅ ✅ ✅ ❌
[Li et al. 2017] ✅ ❌ ✅ ❌ ✅ ✅ ✅ ❌ ❌
SCULPTOR (Ours) ✅ ✅ ✅ ✅ ✅ ✅ ✅ ✅ ✅
Reading NotesResearchFace ModelAbout 2 minAbout 721 words
Phace: Physics-based Face Modeling and Animation

1. Introduction

Contributions

  • a novel muscle activation model that offers more flexibility than standard fiber-based models
  • a physics-based simulation method that retains realism even with significant external forces or substantial modifications of the face geometry and tissue material properties
  • an inverse modeling optimization to adapt the simulation template to a series of expression scans of a specific person.
Reading NotesResearchFace ModelPhysicsAbout 5 minAbout 1526 words
SCAPE: Shape Completion and Animation of People

TL;DR

  • Pose Deformation
    • Input — mesh, align with template, different poses of a human
    • Params — pose deformation ( twist vector )
    • Output – mesh
  • Body Deformation
    • Input — mesh, different individuals
    • Params — body deformation ( PCA )
    • Output – mesh
Reading NotesResearchBody ModelAbout 2 minAbout 642 words
Automatic Rigging and Animation of 3D Characters

TL;DR

  • Pinocchio
    • Input — a static character mesh + a generic skeleton
    • Output — adapt skeleton to character and attach it to surface
  • Skeleton Embedding
    • Discretization
      • Distance Field
      • Approximate Medial Surface
      • Sphere Packing
      • Graph Construction
    • Reduced Skeleton
    • Discrete Penalty Function
    • Embedding Refinement
  • Skin Attachment — attach skeleton to surface
Reading NotesResearchBody ModelAbout 3 minAbout 790 words
Learning a model of facial shape and expression from 4D scans

3. Model formulation

FLAME is described by a function:

M(β⃗,θ⃗,ψ⃗):R\absβ⃗×\absθ⃗×\absψ⃗→R3N M\pqty{\vec{\beta}, \vec{\theta}, \vec{\psi}}: \mathbb{R}^{\abs{\vec{\beta}} \times \abs{\vec{\theta}} \times \abs{\vec{\psi}}} \to \mathbb{R}^{3 N}

Reading NotesResearchFace ModelAbout 5 minAbout 1622 words
A Morphable Model For The Synthesis Of 3D Faces

[PDF] acm.org

TL;DR

  • focus on identity

  • Morphable Model — averages of shape Sˉ\bar{S} and texture Tˉ\bar{T} + eigenvectors sis_i and tit_i

  • Formula — Smod=Sˉ+∑i=1m−1αisiS_{mod} = \bar{S} + \sum_{i = 1}^{m - 1} \alpha_i s_i

  • Building

    • Input — 3D Scans
    • Method — Optic Flow + PCA ( see Sec. 5 )

  • Matching / Register — minimize difference between images / 3D scans

Reading NotesResearchFace ModelAbout 4 minAbout 1316 words