arrow
返回

Robust Motion In-betweening

delete2020-08-12
delete157
delete
OA
AI
F
Félix G. Harvey *
D
Derek Nowrouzezahrai
C
Christopher Pal
DOI:10.1145/3386569.3392480delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
In this work we present a novel, robust transition generation technique that can serve as a new tool for 3D animators, based on adversarial recurrent neural networks. The system synthesises high-quality motions that use temporally-sparse keyframes as animation constraints. This is reminiscent of the job of in-betweening in traditional animation pipelines, in which an animator draws motion frames between provided keyframes. We first show that a state-of-the-art motion prediction model cannot be easily converted into a robust transition generator when only adding conditioning information about future keyframes. To solve this problem, we then propose two novel additive embedding modifiers that are applied at each time-step to latent representations encoded inside the network's architecture. One modifier is a time-to-arrival embedding that allows variations of the transition length with a single model. The other is a scheduled target noise vector that allows the system to be robust to target distortions and to sample different transitions given fixed keyframes. To qualitatively evaluate our method, we present a custom MotionBuilder plugin that uses our trained model to perform in-betweening in production scenarios. To quantitatively evaluate performance on transitions and generalizations to longer time horizons, we present well-defined in-betweening benchmarks on a subset of the widely used Human3.6M dataset and on LaFAN1, a novel high quality motion capture dataset that is more appropriate for transition generation. We are releasing this new dataset along with this work, with accompanying code for reproducing our baseline results.
Keyword:
animation
locomotion
transition generation
in-betweening
deep learning
LSTM
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

ACM Transactions on Graphics 封面图
ACM Transactions on Graphics
IF:
9.5
论文数:
4.7K
被引数:
3.6W

机构

U
ubisoft entertainment
学者数:
8
论文数: 8
被引数: 0
U
universite de montreal
学者数:
4.6W
论文数: 3.8W
被引数: 46
P
Polytechnique Montreal
学者数:
3.7K
论文数: 3.4K
被引数: 42
学者 查看更多机构
引用论文

引用论文

Factors influencing Local Communities Relational Values to Forest Protected Areas in Jordan
err2020-12-01
err0
PREAI
errSalsabeel Shishany; Amani A. Al-Assaf; Mohammad Majdalawi; Mohammad Tabieh; Maher Tadros
err分享
err收藏
err分享
err收藏
Nephrolithiasis in a 17-Year-Old Male With Seckel Syndrome and Horseshoe Kidneys: Case Report and Review of the Literature
err2018-10-01
err0
PREAI
errMatthew Jung; Arun Rai; Lily Wang; Kathleen Puttmann; Kamlesh Kukreja; Chester J. Koh
err分享
err收藏
Long-Term Longitudinal Patterns of Patient-Reported Fatigue After Breast Cancer: A Group-Based Trajectory Analysis
err2022-07-01
err0
errOAAI
errInes Vaz-Luis; Antonio Di Meglio; Julie Havas; Mayssam El-Mouhebb; Pietro Lapidari; Daniele Presti; Davide Soldato; Barbara Pistilli; Agnes Dumas; Gwenn Menvielle; Cecile Charles; Sibille Everhard; Anne-Laure Martin; Paul H. Cottu; Florence Lerebours; Charles Coutant; Sarah Dauchy; Suzette Delaloge; Nancy U. Lin; Patricia A. Ganz; Ann H. Partridge; Fabrice André; Stefan Michiels
err分享
err收藏
Continuous Character Control with Low-Dimensional Embeddings具有低维嵌入的连续字符控制
err2012-07-01
err143
PREAI
errLevine, Sergey; Wang, Jack M.; Haraux, Alexis; Popovic, Zoran; Koltun, Vladlen
err分享
err收藏
Key Lessons From Tailoring Agile Methods for Large-Scale Software Development
err2019-01-01
err0
errOAAI
errTorgeir Dingsoyr; Tore Dyba; Mette Gjertsen; Anette Odgaard Jacobsen; Tor-Erik Mathisen; Jan Ole Nordfjord; Kjetil Roe; Kjetil Strand
err分享
err收藏
A facile electrochemical sensor based on a composite of electrochemically reduced graphene oxide and a PEDOT:PSS modified glassy carbon electrode for uric acid detection
err2022-02-28
err0
errOAAI
errBudi R. Putra; Ulfiatun Nisa; Rudi Heryanto; Eti Rohaeti; Munawar Khalil; Arini Izzataddini; Wulan T. Wahyuni
err分享
err收藏
学者 查看更多内容