融合SPA遮挡分割的多目标跟踪方法  

Multi-target tracking approach combined with SPA occlusion segmentation

作　　者：丁欢[1] 张文生[1] 

机构地区：[1]中国科学院自动化研究所控制国家重点实验室,北京100190

出　　处：《中国图象图形学报》2012年第1期90-98,共9页Journal of Image and Graphics

基　　金：国家自然科学基金项目(90924026);国家高技术研究发展计划(863)项目(2008AA01Z121;2007AA01Z338);首都科技条件平台中国科学院研发实验服务基地研发实验基金项目

摘　　要：复杂环境下的多目标视频跟踪是计算机视觉领域的一个难点,有效处理目标间遮挡是解决多目标跟踪问题的关键。将运动分割方法引入目标跟踪领域,提出一种融合骨架点指派(SPA)遮挡分割的多目标跟踪方法。由底层光流信息得到骨架点,并估计骨架点遮挡状态;综合使用目标外观、运动、颜色信息等高级语义信息,将骨架点指派给各个目标;最后以骨架点为核,对运动前景密集分类,得到准确的目标前景像素;在粒子滤波器跟踪框架下,使用概率外观模型进行多目标跟踪。在PETS2009数据集上的实验结果表明,文中方法能够改进现有多目标跟踪方法对目标间交互适应性较差的缺点,更好地处理动态遮挡问题。Multiple-target tracking in complex scenes is one of the most complicated problems in computer vision. Handling occlusions between objects is the key issue in multiple-target tracking. This paper introduces a method of motion segmentation into the object tracking system, and presents a SPA （skeleton points assign） based occlusion segmentation approach to track multiple people through complex situations which are captured by static monocular cameras. In the proposed method, we select the skeleton points and evaluate their occlusion states by bottom information like optical flow; then we assign these points to different objects using advanced semantic information, such as appearance, motion, and color. Finally a dense classification of foreground pixels is used to accomplish occlusion segmentation. Object tracking is handled by a particle filter-based tracking framework, and a probabilistic appearance model is used to find the best particle. Experiments are performed using the public challenging data set PETS 2009. The experimental results show that our approach can improve the performance of the existing tracking approach and handle dynamic occlusions better.

关 键 词：多目标跟踪 遮挡分割 粒子滤波 骨架点 运动分割 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]