Modeling Human Motion Using Binary Latent Variables
Modeling Human Motion Using Binary Latent Variables
复制标题
DOI：
10.7551/mitpress/7503.003.0173
复制
发表时间：
2006-12
期刊：
影响因子：
--
通讯作者：
Graham W. Taylor;Geoffrey E. Hinton;S. Roweis
中科院分区：
文献类型：
--
作者：
Graham W. Taylor;Geoffrey E. Hinton;S. Roweis
我们提出了一个非线性生成模型的人体运动数据，使用一个无向模型与二进制的潜变量和实值的“可见”变量，代表关节角度。每个时间步的隐变量和可见变量都从最后几个时间步的可见变量接收有向连接。这样的架构使得在线推理效率高，并允许我们使用一个简单的近似学习过程。训练后，模型会找到一组参数，同时捕捉几种不同类型的运动。我们证明了我们的方法的力量，通过合成各种运动序列，并通过执行在线填充在运动捕捉过程中丢失的数据。
We propose a non-linear generative model for human motion data that uses an undirected model with binary latent variables and real-valued "visible" variables that represent joint angles. The latent and visible variables at each time step receive directed connections from the visible variables at the last few time-steps. Such an architecture makes on-line inference efficient and allows us to use a simple approximate learning procedure. After training, the model finds a single set of parameters that simultaneously capture several different kinds of motion. We demonstrate the power of our approach by synthesizing various motion sequences and by performing on-line filling in of data lost during motion capture.