Development of multiple human tracking technique under complex situations by integrating various models

提出了一种新的复杂场景下的多人跟踪方法。利用立体摄像机获取的颜色和距离信息，建立了人体提取和跟踪的随机模型。此外，还建立了基于离散选择模型的行人行为仿真模型。根据上述模型，我们开发了一种方法，作为一个随机过程，集成在一般状态空间模型的框架。我们将所提出的方法应用于在火车站检票口采集的数据，并证实了该方法的高性能。

We propose a new method of multiple human tracking under the complex situations. By using color and range information acquired by stereo video cameras, stochastic models for human extraction and tracking are developed. Additionally, simulation model of pedestrian behavior is also developed based on a discrete choice model. According to the above models, we develop a method as a stochastic process, integrating them on the framework of general state space model. We apply the proposed method to the data acquired at the ticket gate of the railway station and confirm the high performance of the method.