Nonparametric discovery of activity patterns from video collections
Nonparametric discovery of activity patterns from video collections
复制标题
DOI：
10.1109/cvprw.2012.6239170
复制
发表时间：
2012-06
期刊：
影响因子：
--
通讯作者：
M. Hughes;Erik B. Sudderth
中科院分区：
文献类型：
--
作者：
M. Hughes;Erik B. Sudderth
我们提出了一个基于Beta过程的非参数框架，用于发现异质视频集合中的时间模式。我们从量化的局部运动描述符出发，通过一组动态行为之间的转换来描述每个视频的远程时间动态。贝叶斯非参数统计方法允许在没有监督的情况下学习每个视频所展示的此类行为的数量和子集。我们通过两种方式扩展了早期的Beta过程HMM：添加数据驱动的MCMC移动以改进对真实数据集的推理，以及允许行为转移参数的全局共享。我们举例说明了从简单练习、食谱准备和奥林匹克运动的视频中发现直观和有用的动态结构，在各种时间尺度上。分割和检索实验表明了非参数方法的优点。
We propose a nonparametric framework based on the beta process for discovering temporal patterns within a heterogenous video collection. Starting from quantized local motion descriptors, we describe the long-range temporal dynamics of each video via transitions between a set of dynamical behaviors. Bayesian nonparametric statistical methods allow the number of such behaviors and the subset exhibited by each video to be learned without supervision. We extend the earlier beta process HMM in two ways: adding data-driven MCMC moves to improve inference on realistic datasets and allowing global sharing of behavior transition parameters. We illustrate discovery of intuitive and useful dynamical structure, at various temporal scales, from videos of simple exercises, recipe preparation, and Olympic sports. Segmentation and retrieval experiments show the benefits of our nonparametric approach.