Example-Based Synthesis of Stylized Facial Animations

Jakub Fišer
CTU in Prague, FEE
 
Ondřej Jamriška
CTU in Prague, FEE
 
David Simons
Adobe Research
Eli Shechtman
Adobe Research
 
Jingwan Lu
Adobe Research
 
Paul Asente
Adobe Research
Michal Lukáč
Adobe Research
 
Daniel Sýkora
CTU in Prague, FEE

New! Puppetron was announced at Adobe MAX 2017

FaceStyle demo was released at SIGGRAPH 2017



Abstract

We introduce a novel approach to example-based stylization of portrait videos that preserves both the subject's identity and the visual richness of the input style exemplar. Unlike the current state-of-the-art based on neural style transfer [Selim et al. 2016], our method performs non-parametric texture synthesis that retains more of the local textural details of the artistic exemplar and does not suffer from image warping artifacts caused by aligning the style exemplar with the target face. Our method allows the creation of videos with less than full temporal coherence [Ruder et al. 2016]. By introducing a controllable amount of temporal dynamics, it more closely approximates the appearance of real hand-painted animation in which every frame was created independently. We demonstrate the practical utility of the proposed solution on a variety of style exemplars and target videos.

Full Text     Slides     Demo     BibTeX

ACM Transactions on Graphics 36(4):155, 2017

(SIGGRAPH 2017, Los Angeles, USA, July 2017)

Patent pending

Comparison

Style ExemplarOur Result

Style ExemplarOur Result

Style ExemplarOur Result [Selim et al. 2016]

Style ExemplarOur Result [Liao et al. 2017]

Media


 

 

 




=> Back to main page <=