← ArchivePaper2024

A Facial Motion Retargeting Pipeline for Appearance Agnostic 3D Characters

Anatomy-based pipeline translates mocap markers to FACS muscle activations with explicit passive-muscle modeling for diverse character faces.

Open the paper ↗Trace in the lineage graph

Abstract

3D facial motion retargeting has the advantage of capturing and recreating the nuances of human facial motions and speeding up the time‐consuming 3D facial animation process. However, the facial motion retargeting pipeline is limited in reflecting the facial motion's semantic information (i.e., meaning and intensity), especially when applied to nonhuman characters. The retargeting quality heavily relies on the target face rig, which requires time‐consuming preparation such as 3D scanning of human faces and modeling of blendshapes. In this paper, we propose a facial motion retargeting pipeline aiming to provide fast and semantically accurate retargeting results for diverse characters. The new framework comprises a target face parameterization module based on face anatomy and a compatible source motion interpretation module. From the quantitative and qualitative evaluations, we found that the proposed retargeting pipeline can naturally recreate the expressions performed by a motion capture subject in equivalent meanings and intensities, such semantic accuracy extends to the faces of nonhuman characters without labor‐demanding preparations.

How to read this

Category: Method: facial motion retargeting pipeline (anatomy-based)
Contributions: An anatomy-based target-face parameterization module plus a compatible source-motion interpretation module that maps mocap markers to FACS-style muscle activations.
Explicit modeling of passive muscle behavior to preserve semantic meaning and intensity of expressions.
Aims for fast, semantically accurate retargeting to diverse and nonhuman characters without labor-heavy rig preparation (e.g. 3D scanning, blendshape modeling).
Context: Extends artist-friendly facial retargeting (cf. Seol et al., 2011) by grounding the target representation in face anatomy rather than per-character blendshapes.Builds on: Artist Friendly Facial Animation Retargeting
Correctness: Reported via quantitative and qualitative evaluation showing preserved meaning and intensity, including on nonhuman faces; 'semantic accuracy' rests on the anatomy/FACS parameterization being a faithful proxy, and results are evaluation-set dependent rather than a guarantee for arbitrary rigs.
Clarity: Accessible if you know FACS and retargeting; a first pass conveys the two-module structure, a second pass for the passive-muscle modeling and parameterization.
How to read it: Focus on the source-to-target mapping and how semantics (meaning, intensity) are preserved; second pass on the anatomy parameterization if you target nonhuman or appearance-agnostic rigs.

Builds on

Artist Friendly Facial Animation Retargeting 2011

Built upon by

Nothing yet.

Related work

Vision-based control of 3D facial animation 2003 / SCA
Facial Retargeting with Automatic Range of Motion Alignment 2017 / SIGGRAPH
Semi-Supervised Video-Driven Facial Animation Transfer for Production 2021 / SIGGRAPH Asia
Local Anatomically-Constrained Facial Performance Retargeting 2022 / SIGGRAPH

Keywords

facial motion retargeting motion capture appearance agnostic 3d character animation facial animation transfer facial dynamics character neutral model expression transfer

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →