Target tracking method based on particle filtering and depth distance metric learning

The invention discloses a target tracking method based on particle filtering and depth distance metric learning, and relates to the field of automatic driving target visual tracking. The method comprises the following steps: constructing a nonlinear depth metric learning model; training the nonlinear deep metric learning model based on a given automatic driving target positive and negative sample set, and optimizing nonlinear deep learning model parameters based on a gradient descent method; constructing a target observation model based on particle filtering to obtain optimal estimation of an automatic driving target state; and updating the target template through an online tracking strategy combining short-time and long-time stable updating so as to realize effective tracking of the automatic driving target. The method is good in performance under the scenes of partial shielding, illumination variation and the like. Compared with a comparison algorithm, under most test scenes, the method is low in average ce