Vision object tracking method based on hierarchical convolution

Abstract

The present invention provides a vision object tracking method based on hierarchical convolution. The method mainly comprises the content consisting of hierarchical convolution, correlation filters, translation estimation from rough to fine, region proposal and model updating. The method comprises the processes of: employing hierarchical features in a convolutional layer, and employing bilinear interpolation to regulate each feature map to a larger fixed dimension; performing normalization of the cycle version of input features to a soft target score generated by a Gaussian function; searchingthe maximum value of a target object on a response map; giving a related response map set; performing hierarchical deduction of each layer of target translation; calculating one confidence coefficient score of each proposal; keeping long-term memory of the target appearance; and finally, performing minimization of output errors to update an optimal filter. The vision object tracking method basedon hierarchical convolution mitigates sampling fuzziness, reduces tracking drift, reduces errors caused by reasons such as illumination change, shielding, background hybridization, sudden movement andtarget drift out of a visual field, and improves identification accuracy and robustness.

Description

technical field [0001] The invention relates to the field of visual object tracking, in particular to a visual object tracking method based on layered convolution. Background technique [0002] With the improvement of computer performance and the development of vision technology, the tracking and detection of visual objects has attracted more and more attention of researchers. Visual object tracking will have a very broad application prospect in the future: such as computer human-computer interaction, using visual tracking to recognize human gestures and dumb words, so that people with disabilities or working in special working environments can also operate computers; The position, shape and motion speed of the external environment can be used for the navigation of unmanned vehicles or various mobile robots; the use of visual tracking methods to analyze the information and motion parameters of objects in medical images can give doctors key information Remind, assist doctors...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem of decreased accuracy of visual tracking, the object of the present invention is to provide a visual object tracking method based on layered convolution, first utilize the layered features in the convolutional layer, and use bilinear interpolation to convert each feature map Resizing to a larger fixed size, followed by normalizing the recurrent version of the input features to a soft target score produced by a Gaussian function, and searching for the maximum value on the target object response map, and then given the set of relevant response maps, hierarchically infers each One layer of object translation, computing a confidence score for each proposal, maintaining a long-term memory of object appearance, and finally updating the optimal filter by minimizing the output error

Images