VGG deep network-based visual target tracking method

Abstract

The invention discloses a VGG deep network-based visual target tracking method. The method comprises the following steps of S1, compiling a running environment of MatConvNet; S2, establishing a VGG deep neural network; S3, performing video frame input, and judging whether an input frame is an initial frame or not, if the input frame is a non initial frame, entering the step S4, and if the input frame is the initial frame, skipping the step S4 and entering the step S5; S4, performing estimation of a new state of a target, and entering the step S5; and S5, performing online update of a filter model. Compared with a conventional visual target tracking method, more semantic information is comprised in features, and higher tracking precision can be achieved. Compared with a visual target tracking method using high-layer depth features, used low-layer data can reduce the calculation loss and does not lack the semantic information. Therefore, the tracking precision and the tracking speed areweighed, and excellent tracking performance is obtained.

Description

technical field [0001] The invention relates to a visual target tracking method, in particular to a visual target tracking method based on a VGG deep network, and belongs to the technical field of computer vision and visual tracking. Background technique [0002] Visual object tracking is one of the very challenging problems in the field of computer vision. The task of visual object tracking is to estimate the state of the object in subsequent frames given the state of the object in the initial frame (ie, position, size, etc.) in the video sequence. Although the visual target tracking technology has developed rapidly in recent years, due to the influence of factors such as target occlusion, appearance deformation, motion blur, fast motion, illumination changes, scale changes, and complex backgrounds during the tracking process, the application of visual target tracking technology is still difficult. . [0003] Visual object tracking methods are mainly divided into two cate...

AI Technical Summary

Problems solved by technology

The difference between deep features and traditional features is that deep features bring semantic information that traditional features do not have, which is of great significance for enriching feature information and enhancing feature expression capabilities. At the same time, it brings more The calculation loss makes the tracking process difficult to real-time

[0005] To sum up, one of the problems existing in the existing visual target tracking technology is that the quality of feature information greatly affects the performance of tracking, and rich feature information and real-time tracking speed cannot be combined.

Images