Scale adaptive kernel correlation filter fast target tracking method

Abstract

The invention relates to a scale adaptive kernel correlation filter fast target tracking method. According to the scale adaptive kernel correlation filter fast target tracking method, separation of afront background is promoted and a scale estimation module is added through introducing a two-dimensional variable bandwidth Gaussian window into a KCF framework, scale variations of a target are estimated according to a matching result of feature points of an image block, and whether the target is sheltered is judged by combining displacement prediction based on the scale variations. Once the target is sheltered, the user considers that no scale conversion occurs at the moment and template update is not performed, further the tracking precision is enhanced in this way. The scale adaptive kernel correlation filter fast target tracking method has the advantages of simple and fast calculation, and can significantly improve the calculation efficiency on the basis of compatible scale adaptation.

Description

technical field [0001] The invention relates to a Kernel Correlation Filter target tracking technology, in particular to a scale-adaptive fast Kernel Correlation Filter target tracking, which belongs to the technical field of video analysis. Background technique [0002] Object tracking is one of the core problems of video analysis, and it has a wide range of applications in human-computer interaction, video surveillance, augmented reality and other fields. Although the research on this technology has made great progress in the past few decades, due to various factors such as illumination changes, rigid deformation, fast motion, partial occlusion, and complicated backgrounds in the application, it is necessary to seek fast and stable tracking suitable for object changes. Methodology is always a challenge. [0003] In recent years, detection-based tracking (tracking-by-detection) is a typical object tracking mode. This tracking mode integrates online learning with template ...

AI Technical Summary

Problems solved by technology

This method divides color attributes into 11 categories, selects more prominent colors in a timely manner, and uses a dimensionality reduction method similar to PCA to reduce the feature vector from 11 dimensions to 2 dimensions, which improves the accuracy, but for multi-scale targets The tracking performance is not good enough

Another scale adaptive kernel correlation filter tracker based on feature integration [Li Y,Zhu J.A Scale Adaptive Kernel Correlation Filter Tracker with FeatureIntegration[C].European Conference on Computer Vision.Springer,Cham,2014:254-265.], Although this method can solve the problem of the scale change of the target, since it contains 7 predefined scale ratios, it needs to perform bilinear interpolation on the image block every time to adjust the size, which is very complicated and greatly affects the efficiency.

To sum up, the existing KCF improvement method is difficult to take into account both scale adaptation and computational efficiency

Images