赵丽军


摘要:针对目标纹理变化、光照和位置变化较大时,跟踪不稳定、易丢失目标的问题,提出通过多示例学习的训练数据生成局部稀疏编码,建立对象的外观模型。首先,目标对象的局部图像块由过完备字典结合稀疏编码表示;其次,分类器学习稀疏编码进而识别背景中的目标;最后,将训练分类器得到的结果输入粒子滤波框架,进而预测目标状态随时间的变化。此外,为了减少字典更新和分类器累积误差形成的视觉漂移,采用弱分类器结合强分类器进行目标跟踪。
关键词:局部稀疏表示;多示例学习;分类器
中图分类号:TP391 文献标识码:A 文章编号:1009-3044(2015)05-0216-03
Multiple Instance Learning Tracking Method with Local Sparse Representation
ZHAO Li-jun
(China Coal Ping Shuo Outdoor equipment maintenance center, Shuozhou 036006, China)
Abstract:When objects undergo large pose change, illumination variation or partial occlusion, most existed visual tracking algorithms tend to drift away from targets and even fail in tracking them. To address this issue, in this study, the authors propose an online algorithm by combining multiple instance learning (MIL) and local sparse representation for tracking an object. The key idea in our method is to model the appearance of an object by local sparse codes that can be formed as training data for the MIL framework. First, local image patches of a target object are represented as sparse codes with an over complete dictionary. Then MIL learns the sparse codes by a classifier to discriminate the target from the background. Finally, results from the trained classifier are input into a particle filter framework to sequentially estimate the target state over time in visual tracking. In addition, to decrease the visual drift because of the accumulative errors when updating the dictionary and classifier, a two-step object tracking method combining a weak classifier with a strong classifier is proposed.
Key words:local sparse representation; MIL(multiple instance learning);classifier
对象跟踪式计算机视觉领域的一个突出问题。目前正日益广泛应用在视频监控、人机交互、车辆导航等众多领域。然而尽管经过十几年的研究发展,设计一个能够有效处理遮挡、光照变化、尺度变化、运动突变等因素的跟踪算法仍然是一个具有挑战性的问题。
Babenko etal提出的在线MIL算法可以降低目标跟踪中的视觉漂移,但是频繁跟踪新目标需不断更新MIL外观模型,引入的噪声会影响包中的实例的高效选取。为了减少视觉漂移,Matthews etal 提出分两步建立跟踪目标的外观模型。首先采用最新框架中的在线模板估计跟踪结果,然后使用模板确定的第一帧确定目标位置,但是在目标部分遮掩时会有视觉漂移。Wang etal 提出两阶段算法,即同时检测第一帧的背景和在线观察目标。但是,该种方法将目标建模为单一实例,采用稀疏编码建立线性分类器,很难精确选择正负样本更新线性分类器。……