Robust Motion In-betweening
Robust Motion In-betweening
复制标题
DOI：
10.1145/3386569.3392480
复制
发表时间：
2020-07-01
影响因子：
6.2
通讯作者：
Pal, Christopher
中科院分区：
文献类型：
--
作者：
Harvey, Felix G.;Yurick, Mike;Pal, Christopher
在这项工作中，我们提出了一种新的，强大的过渡生成技术，可以作为一个新的工具，为3D动画，基于对抗循环神经网络。该系统合成高质量的运动，使用时间稀疏的关键帧作为动画约束。这让人想起传统动画管道中的中间插入工作，其中动画师在提供的关键帧之间绘制运动帧。我们首先表明，一个国家的最先进的运动预测模型不能很容易地转换成一个强大的过渡生成器时，只添加有关未来的关键帧的条件信息。为了解决这个问题，我们提出了两个新的添加剂嵌入修改器，在每个时间步应用于网络架构内编码的潜在表示。一个修改器是到达时间嵌入，允许单个模型的过渡长度的变化。另一个是预定的目标噪声向量，其允许系统对目标失真具有鲁棒性，并且在给定固定关键帧的情况下对不同过渡进行采样。为了定性地评估我们的方法，我们提出了一个自定义的MotionBuilder插件，该插件使用我们的训练模型在生产场景中执行中间插入。为了定量评估过渡和推广到更长时间范围的性能，我们在广泛使用的Human3.6M数据集的一个子集和LaFAN 1上提供了定义良好的中间基准，LaFAN 1是一种更适合过渡生成的新型高质量运动捕捉数据集。我们将沿着这项工作一起发布这个新的数据集，并附带用于重现我们的基线结果的代码。
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.