← ArchivePaper2020
Interactive Sculpting of Digital Faces Using an Anatomical Modeling Paradigm
Aurel Gruber, Marco Fratarcangeli, Gaspard Zoss, Roman Cattaneo, Thabo Beeler, Markus Gross, Derek Bradley
Interactive face sculpting tool guided by an anatomical model of skull, fat, and tissue, producing physiologically plausible digital faces.
Abstract
Digitally sculpting 3D human faces is a very challenging task. It typically requires either 1) highly‐skilled artists using complex software packages for high quality results, or 2) highly‐constrained simple interfaces for consumer‐level avatar creation, such as in game engines. We propose a novel interactive method for the creation of digital faces that is simple and intuitive to use, even for novice users, while consistently producing plausible 3D face geometry, and allowing editing freedom beyond traditional video game avatar creation. At the core of our system lies a specialized anatomical local face model (ALM), which is constructed from a dataset of several hundred 3D face scans. User edits are propagated to constraints for an optimization of our data‐driven ALM model, ensuring the resulting face remains plausible even for simple edits like clicking and dragging surface points. We show how several natural interaction methods can be implemented in our framework, including direct control of the surface, indirect control of semantic features like age, ethnicity, gender, and BMI, as well as indirect control through manipulating the underlying bony structures. The result is a simple new method for creating digital human faces, for artists and novice users alike.
How to read this
- Category
- Method: an interactive, anatomically guided face-sculpting system
- Contributions
- An anatomical local face model (ALM) of skull, fat, and tissue, learned from several hundred 3D face scans.
- An interactive editing loop where user edits become constraints for a data-driven optimization that keeps results plausible.
- Several interaction modes: direct surface control, indirect semantic control (age, ethnicity, gender, BMI), and control via the underlying bony structures.
- Context
- Builds on anatomically constrained face modeling, notably Wu et al.'s 'An Anatomically Constrained Local Deformation Model for Monocular Face Capture', repurposing that representation from capture toward interactive authoring.Builds on: An Anatomically Constrained Local Deformation Model for Monocular Face Capture
- Correctness
- The plausibility guarantee rests on the scan dataset behind the ALM, so results should be trusted within that population's coverage, and edits far outside the learned distribution may not behave as expected.
- Clarity
- Accessible; a first pass conveys the interaction idea, do a second pass for the ALM optimization formulation.
- How to read it
- Focus first on how edits map to ALM constraints and how the optimization stays plausible; a second pass on the anatomical model and solver pays off if you want to reimplement or extend the editing modes.
Built upon by
Nothing yet.
Related work
Keywords
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 →