Video target tracking method based on twin network fusion multi-template features

Abstract

The invention relates to a video target tracking method based on twin network fusion multi-template features, and provides a semi-supervised template online updating strategy, when a to-be-tracked target in a video sequence has complex conditions such as occlusion, deformation and illumination change, the target change and the occluded condition are evaluated by calculating an APCE value and template similarity, when the appearance of the target is greatly changed, the features extracted from the previous frame of picture are fused with the original template features to obtain a new template with higher expression capability, so that the method is favorable for adapting to various complex conditions; in order to improve the generalization ability of the model and adapt to multiple types oftargets, a regularization technology is adopted in the training process to prevent model overfitting; in order to further improve the speed of the algorithm, only an original template is adopted fortracking in a non-complex situation, so that the calculated amount is greatly reduced, and the method provided by the invention achieves a higher running speed than other methods under the condition of obtaining better tracking performance.

Description

technical field [0001] The invention relates to the field of video target tracking, in particular to a video target tracking method based on twin network fusion multi-template features. Background technique [0002] Video target tracking technology is based on the bounding box information of any object to be tracked given in the first frame of the video sequence, and predicts the position and scale of the bounding box of the same target in subsequent frames. It is widely used in autonomous driving, video surveillance and man-machine areas of interaction. Traditional correlation-based filtering methods use manual features to build filter templates and update them online, such as histogram of oriented gradients (Histogram Of Oriented Gradient, HOG), Haar-like features, and local binary features (Local Binary Pattern, LBP), etc. A series of candidate frames are given first, and then all candidate frames are correlated with the filter template to obtain the confidence of each c...

AI Technical Summary

Problems solved by technology

However, the current target tracking algorithm based on the Siamese network does not have a template update process, or simply replaces the template, resulting in the tracker not being able to adapt well to changes in the appearance of the target.

Images