Gamasutra

If you missed SIGGRAPH 2017 watch a selection of recorded Live Streaming Sessions.

You are here

Technical Papers

Faces & Hair

3D World

HFES

Thursday, 3 August, 2:00 pm - 3:30 pm, Los Angeles Convention Center, Los Angeles Convention Center - Room 153
Session Chair: Ira Kemelmacher-Shlizerman, University of Washington and Facebook

Phace: Physics-Based Face Modeling and Animation

This novel physics-based approach to facial animation computes facial expressions by minimizing a set of non-linear potential energies that model the physical-anatomy interactions. The method enables a variety of advanced animation effects that are difficult to produce using traditional linear-animation techniques.

Alexandru Eugen Ichim
École polytechnique fédérale de Lausanne

Petr Kadlecek
Charles University in Prague, University of Utah

Ladislav Kavan
University of Utah

Mark Pauly
École polytechnique fédérale de Lausanne

Facial Retargeting With Automatic Range-of-Motion Alignment

This approach automatically retargets facial animations even from a real actor to stylized characters. By exploring the similarities of the range of motion defined by the blendshapes of a facial rig and by an expressive training sequence of the actor, it generates actor-personalized blendshapes that are crucial for retargeting.

Roger Blanco Ribera
Korea Advanced Institute of Science and Technology

Eduard Zell
Universität Bielefeld

J.P. Lewis
Victoria University, SEED

Junyong Noh
Korea Advanced Institute of Science and Technology

Mario Botsch
Universität Bielefeld

Example-Based Synthesis of Stylized Facial Animations

This approach to automatically transfer artistic style from an existing exemplar image to a video sequence containing a human face preserves both the subject's identity and the visual richness of the style exemplar, and it introduces a controllable amount of temporal dynamics that approximates the appearance of real hand-painted animations.

Jakub Fišer
Czech Technical University in Prague

Ondřej Jamriška
Czech Technical University in Prague

David Simons
Adobe Research

Eli Shechtman
Adobe Research

Jingwan Lu
Adobe Research

Paul Asente
Adobe Research

Michal Lukáč
Adobe Research

Daniel Sýkora
Czech Technical University in Prague

A Data-Driven Approach to Four-View Image-Based Hair Modeling

Given four input hair images taken from the front, back, left, and right views, this method computes a complete strand-level 3D hair model that closely resembles the hair in all images.

Meng Zhang
Zhejiang University

Menglei Chai
Zhejiang University

Hongzhi Wu
Zhejiang University

Hao Yang
Zhejiang University

Kun Zhou
Zhejiang University