Video-based Toon Character from Surface Performance Capture

彬彬

Introduction

Recent advances in dynamic surface capture have made the creation

of a realistic animation a promising task for modern visual media

production such as computer generated movies or 3D cinematographic

video games. Nonetheless, automatic creation of cartoon

mesh animations, driven by real-life cues, is still a costly and timeconsuming

process and presents a number of hard technical challenges.

To the best of our knowledge, we propose the first attempt

at generating as-photorealistic-as possible cartoon animation from

markerless surface performance capture. The purpose of synthesizing

new puppetry animation demonstrating fidelity to the spirit

of comic book style with more exaggerated motion while preserving

extreme captured cloth wrinkles is dicult to achieve. Consequently,

the key contribution of our work focuses on a novel cartoon

stylization approach for 3D video that eciently reuse temporally

consistent dynamic surface sequence captured from real-world actor

performance. In particular, our simple and eective algorithm

converts realistic spatiotemporal captured surfaces and multi-view

data into an exaggerated life-like squash-and-stretch shape evolution

coupled with context-aware cartoon-style expressive rendering.

2 Our Approach

Inspired by the philosophy adopted for the production of The Adventures

of Tintin: Secret of the Unicorn, an upcoming screen adaptation

including performance capture produced by Weta Digital R

,

we propose an original technique for generating quality life-like

non-photorealistic animation from highly detail animation and photometric

cues captured from real actor performance. Contrary to the

techniques presented in [Kwon and Lee 2011], that only address

the stylization of articulated rigid motion transferred to skinned

meshes, our method generates high-quality cartoon-style 3D video

from real-world dynamic scenes relying on non-rigid underlying

animation structure. Since full-body performance capture is limited

by the physicality of real actors actions, we add more non-natural

squash-and-stretch eects on the global characteristic of photo-real

dynamic surface motions. In addition, original video-driven texture

is not convincing for comic adaptation and invites us to mimic

hand-drawn appearance without the intervention of an artist during

the whole process. To the best of our knowledge, our proposed

method is the first attempt to coherently stylize highly non-rigid

captured surface deformation while preserving photorealistic cues

using non-rigid animation subspace temporal filtering combined

with video-based toon-style shading. As a result, the core algorithm

of our approach is decoupled in two major steps: exaggerating motion

of life-like surface and depicting video-infused appearance.

晃晃

八方捷报，四海增辉。 一元复始，九州同庆； 八方和协，四季平安。 万里江山，增奇添彩； 四化图景，流翠飞红。

晃晃

对你的思念象袅袅的轻烟不绝如缕，对你的祝福是潺潺的小溪叮咚作响。或许岁月将往事褪色，或许空间将彼此隔离。但值得珍惜的依然是你给我的情谊。再次对你说声：新年快乐！

晃晃

好，真棒！！

晃晃

好，真棒！！

奇

加精、加亮滴铁子，尤其要多丁页丁页

奇

其实楼主所说的这些，俺支很少用！

tc

好铁多多发，感激分享

C.R.CAN

佩服，好多阿 ，哈哈

奇

不错 非常经典  实用