In order to improve the tracking accuracy of fast online target tracking and segmentation algorithm, a dynamic weighted siamese network tracking algorithm was proposed. First, the template features extracted from the initial frame and the template features extracted from each frame were learned and fused to improve the generalization ability of the tracker. Second, in the process of obtaining the target mask by the mask branch, the features were fused in a weighting method, so as to reduce the interference caused by redundant features and improve the tracking accuracy. The algorithm was evaluated on the VOT2016 and VOT2018 datasets. The results show that the proposed algorithm has the expected average overlap rate of 0.450 and 0.390 respectively, the accuracy of 0.649 and 0.618 respectively, and the robustness of 0.205 and 0.267 respectively, all of which are higher than those of baseline algorithm. The tracking speed of the proposed algorithm is 34 frame/s, which meets the requirements of real-time tracking. The proposed algorithm effectively improves the tracking accuracy, and completes the tracking task well in a complex tracking environment.

Concerning the real-time performance of the hierarchical convolutional features for visual tracking, a real-time object tracking algorithm based on sparse convolution features was proposed. By analyzing the characteristics of different convolution layers, the equidistant interval sampling method was adopted to extract the sparse convolutional features of each layer. Then the correlation filter response values of each convolutional feature were combined to estimate the location of target object. Finally, a sparse update strategy was applied to improve the computing speed. Experimental results on benchmark dataset OTB-2015 show that the average distance precision of the proposed algorithm is 82.2%, which is 5.25 percentage points higher than that of the original hierarchical convolutional feature tracking algorithm; furthermore, it has better robustness to appearance changes and occlusions. The average tracking speed of the proposed algorithm is 32.6 frames per second, which is nearly 2 times faster than before, which can achieve real-time performance.