Regularized local metric learning for person re-identification

We have proposed a new metric learning approach by exploiting both global and local information of samples to learn similarity for person re-identification.We have presented a regularization approach to address the small sample size problem in previous metric learning based person re-identification approaches.Experimental results on four widely used datasets demonstrate the advantage of our proposed approach over the state-of-the-art methods for person re-identification. In this paper, we propose a regularized local metric learning (RLML) method for person re-identification. Unlike existing metric learning based person re-identification methods which learn a single distance metric to measure the similarity of each pair of human body images, our method combines global and local metrics to represent the within-class and between-class variances. By doing so, we utilize the local distribution of the training data to avoid the overfitting problem. In addition, to address the lacking of training samples in most person re-identification systems, our method also regulates the covariance matrices in a parametric manner, so that discriminative information can be better exploited. Experimental results on four widely used datasets demonstrate the advantage of our proposed RLML over both existing metric learning and state-of-the-art person re-identification methods.