Vision tracking method based on consistency predictor model

Abstract

The invention belongs to the data processing technology field and discloses a vision tracking method based on a consistency predictor model. The method comprises steps that firstly, a two-input convolutional neural network model is constructed, a video frame sampling region and high-level characteristics of a target template are extracted synchronously, and the logistic regression method is utilized to distinguish the target from the background region; secondly, the convolutional neural network is embedded in the consensus predictor framework, the algorithm randomness test is utilized to evaluate reliability of the classification result, under the specified risk level, the classification result with the credibility index is outputted in the domain form; and lastly, the high-confidence region is selected as a candidate target region, through optimizing the space-time domain global energy function, the target trajectory is acquired. The method is advantaged in that the method can be adapted to complex situations such as target blocking, appearance change and background interference, and the method has strong robustness and accuracy compared with presently popular tracking algorithms.

Description

technical field [0001] The invention belongs to the technical field of data processing, in particular to a visual tracking method based on a consistency predictor model. Background technique [0002] Visual object tracking is a basic problem in the field of computer vision. Its task is to determine the motion state of the object in the video, including position, speed and trajectory. Although great progress has been made in visual tracking technology in recent years, it still faces great challenges to achieve robust tracking in complex situations such as object occlusion, pose changes, and mixed backgrounds. [0003] In the visual tracking problem, the feature representation of the target is one of the important factors affecting the tracking performance. The features used to represent objects should be able to adapt to object appearance changes while being well discriminative to background. A large number of feature extraction methods have been applied to visual tracking,...

AI Technical Summary

Problems solved by technology

However, when using deep features for tracking, the common problem is that a large number of samples are required to train and update CNN parameters. For visual tracking tasks, it is usually difficult to obtain a large number of training samples about the tracked target in advance.

[0006] Existing visual tracking methods have poor robustness in target tracking for video sequences; they cannot adapt to complex situations such as target occlusion, appearance changes, and background interference, and the accuracy of various existing tracking algorithms is poor.

Images