Based on the study of point target tracking,the algorithm uses multi-layer convolution feature and correlation filtering technology to track the target.In order to solve the problem that the target is lost in the clutter,the algorithm uses the re-detection mechanism to achieve the long-term target tracking.Firstly,the multi-layer convolution feature of infrared point target is extracted by VGG model.Then the maximum response value is calculated by correlation filtering on each layer.The final response value is obtained by weight fusion to achieve point target tracking.When the target is lost,the re-detection technology is used to detect the target.Then the classifier is updated by using the detection results as labels.Finally,long-term tracking is realized,and theoretical analysis and experimental verification are carried out.The experimental results show that the algorithm is effective and has high accuracy for point target tracking.