中文摘要：

为了提高生成型目标跟踪算法在遮挡、背景干扰等复杂条件下的性能，在稀疏编码模型中引入l0范数正则化约束，以减少冗余编码信息并改善目标表观重构效果。同时提出一种新的基于非凸近端加速梯度的快速迭代算法，解决由此产生的非凸非光滑优化问题。设计了一种增量低秩学习策略，和传统方法需要将目标观测数据作为一个整体进行低秩学习不同，本文方法通过l0正则化稀疏编码能够有效地对目标低秩特征子空间进行在线学习和更新。在多个视频序列上的实验表明：基于l0正则化的增量低秩学习方法能有效提高目标跟踪算法的准确率和鲁棒性；和8种优秀的跟踪算法相比，本文算法在中心误差稳健性和重叠率稳健性两个指标上都取得了最好结果。

英文摘要：

In order to improve the performance of generative visual tracking under complex envaronment such as occlusion and background clutter, firstly,l0 regularized constraint is introduced to sparse coding model to reduce the redundant encoding information and improve the effect of objective apparent recon- struction. As a nontrivial byproduct, a novel fast iterative algorithm based on the non-convex accelerated proximal gradient is proposed to solve the resulting non-convex and non-smooth optimization problems. Secondly,a incremental low-rank features learning strategy is designed. Unlike the traditional methods which need to do low-rank learning on the whole objective observation data matrix,the strategy proposed in this paper can effectively learn and update objective low-rank features subspace online by l0 regular- ized sparse coding. Experimental results on multiple video sequences show that the method based on the l0 regularized incremental low-rank features learning can effectively improve the accuracy and robustness of target tracking algorithm. Compared with the 8 state-of-the-art target tracking algorithms, the pro- posed algorithm achieves the best results both in location error robustness and overlap rate robustness.