中文摘要：

怎样从视频出发准确实时地测量场景中运动人体的高度？针对此问题，本文提出了一种自动实时的人体高度测量方法．该方法首先在视频序列中的每帧图像上提取一种新的头部特征点以及一种新的脚部特征点，然后根据这些特征点建立约束方程求出近似的人体高度，并同时在视频序列中跟踪双脚．最后基于获得的双脚跟踪结果，引入一条关于特征点所对应空间点的几何约束以进一步优化测量结果．与过去的许多测量方法相比，本文方法有效地利用了视频序列中包含的运动信息，有较强的鲁棒性和较高的测量精度，既能有效地处理透视镜头下的视频又能处理鱼眼镜头下的视频，向且计算量很低，可以实现实时测量．实验结果验证了本文算法的有效性和实时性．

英文摘要：

How to measure the height of a moving human from a video sequence？ To solve this problem, a real-time automatic method is proposed. Firstly, a new kind of head feature point and a new kind of feet feature point are extracted in each frame of the input video sequence. Then, these extracted feature points are used to construct several constraint equations to compute the approximate human height, meanwhile the feet are tracked in the video sequence. Finally, based on the tracking results, the measurement result is refined with an introduced geometry constraint on the spatial points corresponding to the extracted feature points. Compared with many previous measurement methods, the proposed method is more accurate and robust because the motion information in the video sequence is utilized effectively, and it can handle the video sequences obtained from both a perspective camera and a fish-eye camera. Moreover, it is a real-time measurement method which has lower computational cost. Experiments demonstrate the effectiveness and the real-time capability of the proposed method.