Target tracking method based on novel visual tracking technology

Abstract

The invention provides a target tracking method based on a novel visual tracking technology. According to the method, a three-dimensional space area is latticed, and is marked with colors or numbers, sublattices to be selected are obtained by moving lattices, the selected sublattices are amplified step by step for visual aids, lattice generation and selection are carried out repeatedly until the set space precision standard is achieved, and finally a target to be selected by a user is determined. According to the method, users of different industries can track targets more visually and vividly according to industry needs or the needs of themselves, target tracking visualization is achieved, the target tracking operation action and state are closely combined, operation is easier and more convenient, the error probability is greatly reduced, and therefore target tracking accuracy is further improved.

Description

technical field [0001] The invention relates to the fields of computer vision and computer graphics, in particular to an object tracking method based on a novel visual tracking technology for object tracking in 3D space. Background technique [0002] At present, when selecting a target at a specific location in 3D space, auxiliary tools are often used to assist users in visual tracking in order to select a suitable target. However, the existing auxiliary tools have certain defects, which prevent users from quickly and accurately selecting the target. Goals are needed for the following reasons: [0003] (1) Spatial sphere-assisted tracking: When the user uses spherical coordinates for target selection, due to the lack of a clear scale, the operator usually needs to continuously rotate or zoom in on the three-dimensional space object to find the target to be selected, which makes the operation more complicated , user tracking is also more difficult; [0004] (2) setting a sc...

AI Technical Summary

Problems solved by technology

[0002] At present, when selecting a target at a specific location in 3D space, auxiliary tools are often used to assist users in visual tracking in order to select a suitable target. However, the existing auxiliary tools have certain defects, which prevent users from quickly and accurately selecting the target. Goals are needed for the following reasons:

[0003] (1) Spatial sphere-assisted tracking: When the user uses spherical coordinates for target selection, due to the lack of a clear scale, the operator usually needs to continuously rotate or zoom in on the three-dimensional space object to find the target to be selected, which makes the operation more complicated , user tracking is also more difficult;

[0004] (2) Set the ruler on the screen, because the set ruler is far away from the content to be selected, it will cause the user's visual separation, so that the user is easy to make a wrong grid when selecting, which increases the error rate of selection

[0005] Due to the lack of 3D visual tracking tools in the prior art, the user often controls the 3D space target selection process manually through the operator, but due to the complicated operation, it is difficult to track human vision, so that the target selection is not enough Intuitive and visual, after setting the ruler on the screen, the distance between the ruler and the content will be far away, and visual separation will occur, and the probability of selection error will increase accordingly, thus affecting the precise positioning of the 3D space target

Images