To address the problem that inaccurate observation of target scale will result in incorrect up-dating of matching template and reduce the accuracy of target tracking in correlation filtering,an ob-ject tracking method assisted by deep snake with correlation filters is proposed.The method consists of two stages.In the first stage,the initial state of target is estimated by the correlation filter.In the sec-ond stage,deep snake is used to get the outline of the target and then optimize its scale.Experimental results on OTB-2015(70%Success)and VOT-2019(28.6%EAO)datasets show that the proposed method has better tracking performance compared with the state-of-the-art methods.