Abstract: This paper presents a novel method of eyes tracking based on a Kalman particle filter. A new approach based on 2D deformable template has been proposed to extract the exact eyes features, then a particle filter is used to track eyes. In order to improve the speed and the precision of the tracking, the paper brings a Kalman filter into the particle filter, which is used to predict and revise in the sampling stage. This method can reduce the number of particles needed in tracking, and realize the purpose that tracks eyes quickly and exactly. Finally, it also performs some experiments, and proves that the method works effectively

Key words: Feature parameters; Kalman particle filter; Eyes tracking

摘要： 提出了一种基于卡尔曼粒子滤波器的人眼跟踪算法，该方法利用一种新的二维可变形模板来提取眼睛的精确特征，采用粒子滤波器跟踪人眼。为了进一步提高普通粒子滤波器跟踪的速度和精度，将卡尔曼滤波器引入粒子滤波器中，利用卡尔曼滤波器算法进行采样预测和校正，减少了人眼跟踪中所需的粒子数目，从而达到快速而准确的跟踪目的。最后，用上述方法进行了实验，验证了该方法的实用性和有效性。

关键词: 特征参数；卡尔曼粒子滤波器；人眼跟踪