基于显著特征区域和条件随机场的目标跟踪

针对目标跟踪中的遮挡问题，提出一种基于局部显著特征区域和条件随机场模型(CRF)的跟踪算法。利用目标区域中的显著特征区域相互之间的空间位置关系以及时间域上相邻区域的影响，并结合各个显著特征区域自身的局部信息建立目标的CRF模型；利用CRF模型对 Mean Shift算法的跟踪结果进行概率推断，融合各个显著特征区域的权重，精确定位运动目标的最终位置。在多个视频序列上的实验结果表明，与改进的 MS算法、粒子滤波算法以及分块跟踪方法相比，文中算法具有较高的跟踪精度；尤其是当目标被遮挡时，该算法具有较好的跟踪鲁棒性。该算法充分利用了显著特征区域自身的局部特征和区域之间的空间结构信息以及各个显著特征区域在时间域上的约束条件，能够实现复杂情况下的运动目标的鲁棒跟踪。

Salient feature regions based conditional random field for object tracking

Occlusion is one of the most challenging problems in obj ect tracking community.To deal with the occlusion problem,this paper presents a local salient feature based probabilistic graphical model for visual tracking.Combining spatial and temporal constraints among different ROIs and local information contained in each ROI,the obj ect is represented as a probabilistic graphical model.Finally,based on the obj ect model and Mean Shift tracking results of each ROI,Probabilistic inference algorithm is adopted to estimate the probability of each ROI belongs to obj ect region. Comprehensive experiments on several testing videos show,compared with three well-known trackers,i.e.improved Mean Shift,particle filter and fragments-based tracker,the proposed method has a higher tracking accuracy and robustness,especially in obj ect occlusion condition.The proposed tracker,using local salient information and spatial and temporal structure constraint of tracking obj ect effectively,can perform with high robustness in complex real-world scenarios such as obj ect occlusion,changes of posture and illumination etc.