Target Tracking Method, Device, Electronic Device and Storage Medium

Abstract

The present disclosure relates to a target tracking method, apparatus, electronic device, and storage medium. The target tracking method includes: obtaining a scale set and an angle set of a target to be tracked in a video sequence, wherein the scale set includes one or more scale parameters that measure the degree of scale change of the target to be tracked, and the angle set includes one or more angle parameters that measure the degree of angular change of the target to be tracked; extracting features from the current video frame based on one or more scale-angle parameter combinations determined by the scale set and the angle set, and obtaining one or more feature maps under different scale-angle parameter combinations; and obtaining the scale and angle of the target to be tracked in the current video frame based on the one or more feature maps, the feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set. The target tracking method, apparatus, electronic device, and storage medium disclosed herein can improve the accuracy of target tracking.

Description

Technical Field

[0001] The present disclosure relates to the field of computer vision technology, and more specifically, to a target tracking method, device, electronic device, and storage medium. Background Art

[0002] Video object tracking is an important research direction in the field of computer vision. Its applications include video surveillance, detection and recognition, human-computer interaction, and unmanned driving. It can also be used as a video editing tool and is widely used in daily production and life.

[0003] At present, most video target tracking methods focus on tracking target displacement (i.e., tracking the position change of a specific target in the picture). However, due to the scale changes and morphological differences of the target itself, simply tracking the target displacement may lead to target loss, and the accuracy of video target tracking is low. Summary of the Invention

[0004] The present disclosure provides a target tracking method, device, electronic device and storage medium to at least solve the problems in the above-mentioned related technologies.

[0005] According to a first aspect of an embodiment of the present disclosure, a target tracking method is provided, comprising: obtaining a scale set and an angle set of a target to be tracked in a video sequence, the scale set including one or more scale parameters that measure the degree of scale change of the target to be tracked, and the angle set including one or more angle parameters that measure the degree of angular change of the target to be tracked; extracting features from a current video frame based on one or more scale-angle parameter combinations determined by the scale set and the angle set, to obtain one or more feature maps of the current video frame under different scale-angle parameter combinations; and obtaining the scale and angle of the target to be tracked in the current video frame based on the one or more feature maps, feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set.

[0006] Optionally, the one or more scale-angle parameter combinations determined according to the scale set and the angle set are used to extract features from the current video frame to obtain one or more feature maps of the current video frame under different scale-angle parameter combinations, including: adjusting the scale angle of the target to be tracked in the previous video frame of the current video frame according to each scale angle parameter combination in the one or more scale angle parameter combinations to obtain one or more adjusted scale angles of the target to be tracked in the current video frame; and extracting features from the current video frame including the target to be tracked with the adjusted scale angle to obtain the one or more feature maps.

[0007] Optionally, obtaining the scale and angle of the target to be tracked in the current video frame based on the one or more feature maps, the feature maps of a preset number of video frames before the current video frame, the scale set and the angle set includes: obtaining a first response based on the feature maps of a preset number of video frames before the current video frame, the scale set and the angle set; obtaining one or more second responses based on the one or more feature maps and the first response, each second response reflecting the degree of proximity between the adjusted scale angle of the target to be tracked in the current video frame and the true scale angle; and determining the adjusted scale angle corresponding to the second response that meets the preset conditions among the one or more second responses as the scale and angle of the target to be tracked in the current video frame.

[0008] Optionally, obtaining the first response based on the feature maps of a preset number of video frames before the current video frame, the scale set and the angle set includes: obtaining a scale Gaussian response and an angle Gaussian response according to the number of scale parameters in the scale set and the number of angle parameters in the angle set, respectively; and obtaining the first response based on the scale Gaussian response, the angle Gaussian response and the feature maps of the two video frames before the current video frame.

[0009] Optionally, the feature maps of the first two video frames of the current video frame include the first feature map and the second feature map of the first video frame before the current video frame and the third feature map of the second video frame before the current video frame; obtaining the first response based on the scale Gaussian response, the angle Gaussian response and the feature maps of the first two video frames of the current video frame includes: performing fast Fourier transform on the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map and the third feature map to obtain the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map and the third feature map in the frequency domain; correlating the scale Gaussian response in the frequency domain with the angle Gaussian response in the frequency domain to obtain the scale-angle comprehensive Gaussian response in the frequency domain; obtaining the first response based on the first feature map, the second feature map, the third feature map and the scale-angle comprehensive Gaussian response in the frequency domain.

[0010] Optionally, obtaining the first response based on the first feature map, the second feature map, the third feature map and the scale-angle integrated Gaussian response in the frequency domain includes: associating the second feature map in the frequency domain and the third feature map in the frequency domain with the scale-angle integrated Gaussian response in the frequency domain, respectively, to obtain a target second feature map in the frequency domain and a target third feature map in the frequency domain, wherein the target second feature map in the frequency domain and the target third feature map in the frequency domain respectively carry scale-angle information; obtaining the first response according to the target second feature map in the frequency domain, the target third feature map in the frequency domain and the first feature map in the frequency domain.

[0011] Optionally, obtaining one or more second responses based on the one or more feature maps and the first response includes: performing a fast Fourier transform on the one or more feature maps to obtain one or more feature maps in the frequency domain; and obtaining the one or more second responses based on each feature map in the one or more feature maps in the frequency domain and the first response.

[0012] Optionally, after determining the adjusted scale angle corresponding to the second response that meets the preset conditions in the one or more second responses as the scale and angle of the target to be tracked in the current video frame, the method also includes: updating the first response based on the feature map of the current video frame and the feature map of the previous video frame of the current video frame.

[0013] According to a second aspect of an embodiment of the present disclosure, a target tracking device is provided, including: a set acquisition unit, configured to: acquire a scale set and an angle set of a target to be tracked in a video sequence, the scale set including one or more scale parameters measuring the degree of scale change of the target to be tracked, and the angle set including one or more angle parameters measuring the degree of angle change of the target to be tracked; a feature extraction unit, configured to: extract features for a current video frame according to one or more scale-angle parameter combinations determined by the scale set and the angle set, and obtain one or more feature maps of the current video frame under different scale-angle parameter combinations; a scale and angle acquisition unit, configured to: obtain the scale and angle of the target to be tracked in the current video frame according to the one or more feature maps, feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set.

[0014] Optionally, the feature extraction unit can be configured to adjust the scale angle of the target to be tracked in the previous video frame of the current video frame according to each scale angle parameter combination in the one or more scale angle parameter combinations, to obtain one or more adjusted scale angles of the target to be tracked in the current video frame; and perform feature extraction on the current video frame including the target to be tracked with adjusted scale angle to obtain the one or more feature maps.

[0015] Optionally, the scale and angle acquisition unit can be configured to obtain a first response based on the feature maps of a preset number of video frames before the current video frame, the scale set and the angle set; obtain one or more second responses based on the one or more feature maps and the first response, each second response reflecting the degree of proximity between the adjusted scale angle of the target to be tracked in the current video frame and the actual scale angle; and determine the adjusted scale angle corresponding to the second response that meets the preset conditions among the one or more second responses as the scale and angle of the target to be tracked in the current video frame.

[0016] Optionally, the scale and angle acquisition unit can be configured to obtain a scale Gaussian response and an angle Gaussian response, respectively, based on the number of scale parameters in the scale set and the number of angle parameters in the angle set; and obtain the first response based on the scale Gaussian response, the angle Gaussian response and the feature maps of the first two video frames of the current video frame.

[0017] Optionally, the feature maps of the first two video frames of the current video frame include the first feature map and the second feature map of the first video frame before the current video frame and the third feature map of the second video frame before the current video frame; the scale and angle acquisition unit can be configured to perform a fast Fourier transform on the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map and the third feature map to obtain the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map and the third feature map in the frequency domain; associate the scale Gaussian response in the frequency domain with the angle Gaussian response in the frequency domain to obtain the scale-angle comprehensive Gaussian response in the frequency domain; and obtain the first response based on the first feature map, the second feature map, the third feature map and the scale-angle comprehensive Gaussian response in the frequency domain.

[0018] Optionally, the scale and angle acquisition unit can be configured to associate the second feature map of the frequency domain and the third feature map of the frequency domain with the scale-angle integrated Gaussian response of the frequency domain, respectively, to obtain the target second feature map of the frequency domain and the target third feature map of the frequency domain, wherein the target second feature map of the frequency domain and the target third feature map of the frequency domain respectively carry scale angle information; and obtain the first response based on the target second feature map of the frequency domain, the target third feature map of the frequency domain and the first feature map of the frequency domain.

[0019] Optionally, the scale and angle acquisition unit can be configured to perform a fast Fourier transform on the one or more feature maps to obtain one or more feature maps in the frequency domain; and obtain the one or more second responses based on each feature map in the one or more feature maps in the frequency domain and the first response.

[0020] Optionally, the target tracking device also includes a first response updating unit, which can be configured to update the first response according to the feature map of the current video frame and the feature map of the previous video frame of the current video frame after determining the adjusted scale angle corresponding to the second response that meets the preset conditions in the one or more second responses as the scale and angle of the target to be tracked in the current video frame.

[0021] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: at least one processor; and at least one memory storing computer-executable instructions, wherein the computer-executable instructions, when executed by the at least one processor, prompt the at least one processor to execute the target tracking method according to the present disclosure.

[0022] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium storing instructions is provided, which, when executed by at least one processor, prompts the at least one processor to execute the target tracking method according to the present disclosure.

[0023] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, wherein instructions in the computer program product can be executed by a processor of a computer device to complete the target tracking method according to the present disclosure.

[0024] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects:

[0025] According to the target tracking method, device, electronic device and storage medium disclosed herein, feature extraction is performed on the current video frame based on the scale set and angle set of the target to be tracked in the video sequence, and one or more feature maps of the current video frame under different scale-angle parameter combinations can be obtained. Based on the obtained one or more feature maps, the feature map, scale set and angle set of the video frame before the current video frame, the scale and angle of the target in the current video frame can be tracked, thereby improving the accuracy of video target tracking.

[0026] In addition, according to the target tracking method, device, electronic device and storage medium disclosed in the present invention, any video target displacement tracking method can be easily superimposed to achieve four-dimensional tracking of the target (i.e., two-dimensional displacement, one-dimensional scale size and one-dimensional angle posture).

[0027] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0029] Figure 1 is a flowchart illustrating a target tracking method according to an exemplary embodiment of the present disclosure.

[0030] Figure 2 is a schematic diagram illustrating a video target tracking process in a specific application scenario according to an exemplary embodiment of the present disclosure.

[0031] Figure 3 FIG. 4 is a schematic diagram illustrating a tracking result of a target tracking method according to an exemplary embodiment of the present disclosure.

[0032] Figure 4 is a block diagram illustrating a target tracking apparatus according to an exemplary embodiment of the present disclosure.

[0033] Figure 5 is a block diagram of an electronic device 500 according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0034] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0035] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation methods described in the following examples do not represent all implementation methods consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0036] It should be noted that the phrase "at least one of the items" in this disclosure includes three types of parallel situations: "any one of the items", "a combination of any multiple items of the items", and "all of the items". For example, "including at least one of A and B" includes the following three parallel situations: (1) including A; (2) including B; (3) including A and B. For another example, "performing at least one of step 1 and step 2" includes the following three parallel situations: (1) performing step 1; (2) performing step 2; and (3) performing steps 1 and 2.

[0037] In order to improve the accuracy and comprehensiveness of video target tracking, the present disclosure proposes a target tracking method, device, electronic device and storage medium. Specifically, based on the scale set and angle set of the target to be tracked in the video sequence, the feature extraction of the current video frame is performed, and one or more feature maps of the current video frame under different scale angle parameter combinations can be obtained. Based on the obtained one or more feature maps, the feature map of the video frame before the current video frame, the scale set and the angle set, the scale and angle of the target in the current video frame can be tracked, thereby improving the accuracy of video target tracking. In addition, any video target displacement tracking method can be easily superimposed on the target tracking method of the present disclosure to achieve four-dimensional tracking of the target (i.e., two-dimensional displacement, one-dimensional scale size and one-dimensional angle posture). Below, reference will be made to Figures 1 to 5 The target tracking method, apparatus, electronic device, and storage medium according to exemplary embodiments of the present disclosure are described in detail.

[0038] First of all, it should be noted that the target tracking method shown in the embodiment of the present disclosure can be used to track video targets. In one embodiment, the target tracking method shown in the present disclosure can be executed by a computer, a laptop, a smart phone, a tablet computer, a wearable device, a vehicle-mounted device, etc. In another embodiment, the target tracking method shown in the present disclosure can also be executed by a chip with computing capabilities, or, in response to a request for tracking video targets sent by the user terminal, it can be executed by a server, a server cluster, a distributed subsystem, a cloud processing platform, a server including a blockchain node, and a combination of devices thereof. The present disclosure does not limit the specific execution device.

[0039] Figure 1 is a flowchart illustrating a target tracking method according to an exemplary embodiment of the present disclosure.

[0040] Reference Figure 1 In step 101, a scale set and an angle set of the target to be tracked in a video sequence are obtained. The scale set includes one or more scale parameters that measure the degree of scale change of the target to be tracked, and the angle set includes one or more angle parameters that measure the degree of angular change of the target to be tracked. Here, the video sequence includes multiple video frames, and the target to be tracked is a moving object in the video sequence. Therefore, the position, scale, and angle of the target to be tracked may vary in different video frames.

[0041] According to an exemplary embodiment of the present disclosure, the scale set may be represented as S={s1,s2,...,s n}, where n is the size of the scale set, s1,s2,…,s n They represent different scale parameters, which are used to measure the scale change of the target to be tracked in the video sequence. In some embodiments, the scale step size can be set to p, and the trackable scale size range is [p*s1, p*s n ], for example, for a scale set S = {0.6, 0.7, 0.8, 0.9, 1, 1.1, 1.2, 1.3, 1.4}, 0.6 means reducing the scale of the target to be tracked to 60% of the original, 1 means the scale of the target to be tracked remains unchanged, and if the scale step p is 0.5, the trackable scale range is [0.3, 0.7]. Therefore, the trackable scale range can be adjusted by changing the scale step without resetting the scale set. The angle set can be expressed as R = {r1, r2, …, r m}, m is the size of the angle set, r1, r2, ..., r m They represent different angle parameters, which are used to measure the degree of angle change of the target to be tracked in the video sequence. In some embodiments, the angle step size can be set to q, and the trackable angle range is [q*r1,q*r m], for example, for the angle set R = {-2, -1, 0, 1, 2}, positive and negative represent two relative directions of angle change (for example, clockwise rotation and counterclockwise rotation, etc.), assuming that the angle step q is 5°, the trackable angle range is [-10°, 10°]. For example, -10° means rotating the angle of the target to be tracked by 10° counterclockwise, 0 means the angle remains unchanged, and 10° means rotating the angle of the target to be tracked by 10° clockwise. In addition, the trackable angle range can be adjusted by changing the angle step. In addition, the specific values of n and m can be determined according to actual conditions (for example, the content type of the video sequence, etc.). The present disclosure does not limit the size of the scale set and the angle set.

[0042] In step 102, based on one or more scale-angle parameter combinations determined by the acquired scale set and angle set, features are extracted for the current video frame to obtain one or more feature maps of the current video frame under different scale-angle parameter combinations. In one embodiment, the feature is a Histogram of Gradient (HOG) feature of the current video frame. Any method in the relevant technology can be used to extract the HOG feature, which is not limited in this disclosure. For example, the current video frame can be divided into a grid of size c*c (e.g., 8*8), a gradient histogram feature is calculated for each grid, and the gradient histogram features of each grid are connected in series to obtain the gradient histogram feature of the current video frame.

[0043] According to an exemplary embodiment of the present disclosure, the scale angle of the target to be tracked in the video frame before the current video frame can be adjusted according to each scale angle parameter combination in one or more scale angle parameter combinations to obtain the adjusted scale angles of one or more targets to be tracked in the current video frame, and feature extraction is performed on the current video frame including the target to be tracked with the adjusted scale angles, thereby obtaining one or more feature maps. Here, the scale and angle of the target to be tracked are adjusted. Specifically, a scale parameter and an angle parameter can be selected from the scale size range pointed to by the scale set and the angle size range pointed to by the angle set as a scale angle parameter combination, and the scale and angle of the target to be tracked in the video frame before the current video frame are transformed according to the scale angle parameter combination. Then, feature extraction is performed on the current video frame including the target to be tracked with the transformed scale and angle, thereby obtaining a feature map under the scale angle parameter combination. In this way, multiple feature maps of the current video frame can be obtained according to different angle parameter combinations. For example, if the scale set and angle set include 3 scale parameters and 3 angle parameters respectively, there are 9 scale angle parameter combinations, and 9 HOG features related to the current video frame can be calculated. Here, different scale angle parameter combinations can be represented by "i, j", and the HOG features of the current video frame under different scale angle parameter combinations can be represented by H i,j Indicates that "i" represents each scale parameter s in the aforementioned scale set n The subscript "j" represents each angle parameter r in the aforementioned angle set. m subscript, that is, 1≤i≤n, 1≤j≤m.

[0044] In step 103, the scale and angle of the target to be tracked in the current video frame are obtained based on the obtained one or more feature maps, the feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set.

[0045] According to an exemplary embodiment of the present disclosure, in order to accurately obtain the scale and angle of the target to be tracked in the current video frame, the scale information and angle information of the target to be tracked in a preset number of video frames before the current video frame can be referenced. Specifically, a first response can be obtained based on the feature map (e.g., gradient histogram HOG), scale set, and angle set of the preset number of video frames before the current video frame, and one or more second responses are obtained based on the obtained one or more feature maps of the current video frame and the first response. Here, the second response reflects the degree of proximity between the adjusted scale angle of the target to be tracked in the current video frame and the actual scale angle. The adjusted scale angle corresponding to the second response (e.g., the second response with the largest value, etc.) that meets the preset conditions in the one or more second responses can be determined as the scale and angle of the target to be tracked in the current video frame.

[0046] Specifically, the scale Gaussian response and the angle Gaussian response can be obtained according to the number of scale parameters in the scale set (for example, the size n of the aforementioned scale set) and the number of angle parameters in the angle set (for example, the size m of the aforementioned angle set), respectively. Here, the scale Gaussian response, for example, but not limited to, can be expressed as G1 = {g1, g2, ..., g n},in, δ1 is a Gaussian parameter greater than 0, e is a natural constant, 1≤i≤n; the angular Gaussian response, for example, but not limited to, can be expressed as G2={h1,h2,…,h m},in, δ2 is a Gaussian parameter greater than 0, 1≤j≤m, δ1 and δ2 can be set according to the actual target tracking scenario, and the present disclosure does not impose any restrictions on this. In addition, in order to accurately obtain the scale and angle of the target to be tracked in the current video frame while minimizing the execution complexity of the target tracking method, the feature maps of the two video frames before the current video frame can be obtained. In some embodiments, the feature maps of the two video frames before the current video frame include a first feature map and a second feature map of the first video frame before the current video frame and a third feature map of the second video frame before the current video frame. Here, the difference between the first feature map and the second feature map is that in the process of obtaining the first feature map and the second feature map, the scale and angle of the target to be tracked may be different. Specifically, for the clarity of description, assuming that the first video frame before the current video frame is video frame A, and the second video frame before the current video frame is video frame B, then in the process of obtaining the first feature map, the scale and angle of the target to be tracked in video frame A are the same as the scale and angle in video frame B, respectively; in the process of obtaining the second feature map, the scale and angle of the target to be tracked in video frame A are the actual scale and actual angle in video frame A; in the process of obtaining the third feature map, the scale and angle of the target to be tracked in video frame B are the actual scale and actual angle in video frame B.

[0047] The first response can be obtained based on the scale Gaussian response, the angle Gaussian response and the feature maps of the first two video frames of the current video frame. Specifically, in order to reduce the amount of calculation during the execution of the target tracking method disclosed in the present invention, the first response can be obtained in the frequency domain, that is, the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map and the third feature map can be fast Fourier transformed to obtain the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map and the third feature map in the frequency domain, and then the scale Gaussian response in the frequency domain and the angle Gaussian response in the frequency domain are associated (for example, the scale Gaussian response in the frequency domain and the angle Gaussian response in the frequency domain are multiplied, etc.) to obtain the scale angle integrated Gaussian response in the frequency domain, and the first response is obtained based on the first feature map, the second feature map and the third feature map in the frequency domain and the scale angle integrated Gaussian response in the frequency domain. In some embodiments, the second characteristic graph of the frequency domain and the third characteristic graph of the frequency domain can be associated with the scale-angle integrated Gaussian response of the frequency domain (for example, the second characteristic graph of the frequency domain and the third characteristic graph of the frequency domain are multiplied with the scale-angle integrated Gaussian response of the frequency domain, etc.) to obtain the target second characteristic graph of the frequency domain and the target third characteristic graph of the frequency domain, where the target second characteristic graph of the frequency domain and the target third characteristic graph of the frequency domain respectively carry scale angle information (that is, carry scale information related to the aforementioned scale set and angle information related to the aforementioned angle set). The first response can be obtained based on the target second characteristic graph of the frequency domain and the target third characteristic graph of the frequency domain and the first characteristic graph of the frequency domain. In one embodiment, different weights can be first superimposed on the target second characteristic graph of the frequency domain and the target third characteristic graph of the frequency domain to determine the contribution of the target second characteristic graph of the frequency domain and the target third characteristic graph of the frequency domain to the first response, and then the target second characteristic graph of the frequency domain and the third characteristic graph of the frequency domain superimposed with different weights are added and divided by the first characteristic graph of the frequency domain to obtain the first response. That is, the first response, for example, but not limited to, can be expressed as:

[0048]

[0049] Among them, F(G1) and F(G2) represent the scale Gaussian response in the frequency domain and the angle Gaussian response in the frequency domain respectively; F(H a ) represents the third feature map of the second video frame before the current video frame in the frequency domain; F(H c ) represents the second feature map of the first video frame before the current video frame in the frequency domain; F(H b) represents the first feature map of the first video frame before the current video frame in the frequency domain; η and (1-η) represent different weights, 0<η<1. It should be noted that if the current video frame is the second video frame in the video sequence, since there is only the first video frame before the second video frame, there is only one reference video frame for the current video frame, and the scale and angle of the target in the first video frame of the video sequence are known, so F(H) is c ) and F(H b ), therefore, for the current video frame being the second video frame of the video sequence, its first response can be expressed as:

[0050]

[0051] At this time, F(H b ) represents the feature map of the first video frame of the video sequence in the frequency domain.

[0052] In another embodiment, different weights may not be superimposed on the target second feature map in the frequency domain and the target third feature map in the frequency domain, and the first response may be obtained directly based on the first feature map in the frequency domain, the target second feature map and the target third feature map. The present disclosure does not impose any restrictions on this.

[0053] According to an exemplary embodiment of the present disclosure, a fast Fourier transform may be performed on one or more feature maps related to the current video frame to obtain one or more feature maps in the frequency domain, and one or more second responses may be obtained based on each feature map and a first response in the one or more feature maps in the frequency domain (for example, each feature map may be multiplied by the first response, etc.).

[0054] Here, the second response, for example, but not limited to, may be expressed as:

[0055] R2=F(H i,j )*R1 (3)

[0056] Among them, “i, j” represents the different scale angle parameter combinations mentioned above, F(H i,j ) represents multiple feature maps of the frequency domain of the aforementioned current video frame under different scale-angle parameter combinations, 1≤i≤n, 1≤j≤m, n is the size of the aforementioned scale set; m is the size of the aforementioned angle set; R1 represents the first response.

[0057] In a specific embodiment, the scale angle parameter combination "i', j'" corresponding to the second response with the largest value among one or more second responses can be recorded, and the scale angle parameter combination "i', j'" can be selected from the aforementioned scale size range [p*s1, p*s n ] and the angle range [q*r1,q*r m] respectively determine the scale adjustment amplitude p*s of the target to be tracked i′ and angle adjustment amplitude q*r j′ , according to the determined scale adjustment amplitude, angle adjustment amplitude and the scale and angle of the target to be tracked in the previous video frame of the current video frame, the scale and angle of the target to be tracked in the current video frame can be obtained. For example, the current video frame is the second video frame in the video sequence, the scale of the target to be tracked in the first video frame of the video sequence is (w1,h1), and the angle is a1, then the scale of the target to be tracked in the second video frame is (w1*p*s i′ ,h1*p*s i′ ), angle a2=a1+q*r j′ .

[0058] According to an exemplary embodiment of the present disclosure, after determining the adjusted scale angle corresponding to the second response that meets the preset conditions in one or more second responses as the scale and angle of the target in the current video frame, the first response can be updated according to the feature map of the current video frame and the feature map of the previous video frame of the current video frame to determine the scale and size of the target to be tracked in the next video frame. For example, with respect to the aforementioned formula (1), when updating the first response R1, F(H c ) and F(H b ) is updated to the second feature map and the first feature map of the current video frame, and F(H a ) is updated to the feature map of the previous video frame of the current video frame.

[0059] The above-mentioned target tracking method can realize the joint tracking of the scale and angle of the video target, and can be easily combined with any displacement tracking method in the related technology to realize four-dimensional tracking of the target (i.e., two-dimensional displacement, one-dimensional scale size and one-dimensional angle posture), thereby improving the comprehensiveness and accuracy of target tracking.

[0060] Based on a combination of one or more of the foregoing embodiments, Figure 2 A schematic diagram showing a video target tracking process in a specific application scenario.

[0061] Reference Figure 2, set the scale set S and the angle set R. The size, step size, and limited scale size range and angle size range of the two can refer to the aforementioned related descriptions and will not be repeated here. According to the scale set S and the angle set R, respectively obtain the scale Gaussian response G1 and the angle Gaussian response G2 (the specific expressions of the two can refer to the aforementioned related descriptions), and perform fast Fourier transform on the scale Gaussian response G1 and the angle Gaussian response G2, respectively, to obtain the scale Gaussian response F(G1) in the frequency domain and the angle Gaussian response F(G2) in the frequency domain. The scale and angle of the target to be tracked in the first video frame of the video sequence are known, where the scale is (w1, h1) and the angle is a1. Extract the HOG feature H1 of the first video frame, and perform fast Fourier transform on the HOG feature H1 to obtain the HOG feature F(H1) in the frequency domain. The first response can be calculated based on F(G1), F(G2), and F(H1). For the second video frame of the video sequence, the scale range defined by the scale set S and the angle range defined by the angle set R can be traversed to calculate the gradient histogram feature H of the second video frame under each scale angle parameter combination "i, j". i,j , and for each H i,j Perform fast Fourier transform to obtain the gradient histogram feature F(H) of the second video frame under each scale angle parameter combination in the frequency domain i,j ), for any scale angle parameter combination "i, j", the second response R2=F(H i,j )*R1, record the scale angle parameter combination "i', j'" when the value of the second response is the largest, and select the scale angle parameter combination "i', j'" from the aforementioned scale size range [p*s1, p*s n ] and the angle range [q*r1,q*r m ] respectively determine the scale adjustment amplitude p*s of the target to be tracked i′ and angle adjustment amplitude q*r j′ , then the scale of the target to be tracked in the second video frame is (w1*p*s i′ ,h1*p*s i′ ), angle a2=a1+p*r j′ . Afterwards, the first response can be updated to Among them, F(H i′,j′) is a feature map obtained by extracting features of the second video frame in the frequency domain according to the actual scale and actual angle of the target to be tracked in the second video frame, F(H2) is a feature map obtained by extracting features of the second video frame in the frequency domain according to the scale and angle of the target to be tracked in the first video frame, and for subsequent video frames, the corresponding second response is obtained according to the same logic as the operation for the second video frame to obtain the scale and angle of the target to be tracked in the subsequent video frames, and the first response R1 is updated until the tracking task of the target to be tracked in all video frames of the video sequence is completed.

[0062] Figure 3 FIG. 4 is a schematic diagram illustrating a tracking result of a target tracking method according to an exemplary embodiment of the present disclosure.

[0063] Reference Figure 3 , Figure 3 (a) is the first video frame of the video sequence, and the position and size of the target 301 to be tracked in the first video frame can be represented by a black solid rectangular box. Figure 3 (b) is the second video frame of the video sequence. Due to factors such as motion, the position of the target to be tracked 301 in the second video moves from the horizontal position of the road section to the slope position, the scale changes from large to small, and the angle between it and the horizontal plane also changes. At this time, it is very likely that the target to be tracked 301 will be lost or the scale and angle of the target to be tracked 301 cannot be tracked. For example, the tracking result obtained using the relevant technology can be represented by a black dotted rectangular box, which can only obtain the position of the target to be tracked in the second video frame, but cannot obtain the size and angle of the target to be tracked 301. However, the target tracking method shown in the present disclosure can track the scale and angle of the target (represented by the black solid line), thereby improving the accuracy of video target tracking.

[0064] Figure 4 is a block diagram illustrating a target tracking apparatus according to an exemplary embodiment of the present disclosure.

[0065] Reference Figure 4 According to an exemplary embodiment of the present disclosure, a target tracking apparatus 400 may include a set acquisition unit 401 , a feature extraction unit 402 , and a scale and angle acquisition unit 403 .

[0066] The set acquisition unit 401 can acquire a scale set and an angle set of the target to be tracked in a video sequence. The scale set includes one or more scale parameters that measure the degree of scale change of the target to be tracked, and the angle set includes one or more angle parameters that measure the degree of angular change of the target to be tracked. Here, the video sequence includes multiple video frames, and the target to be tracked is a moving object in the video sequence. Therefore, the position, scale, and angle of the target to be tracked may vary in different video frames.

[0067] According to an exemplary embodiment of the present disclosure, the scale set may be represented as S={s1,s2,...,s n}, where n is the size of the scale set, s1,s2,…,s n They represent different scale parameters, which are used to measure the scale change of the target to be tracked in the video sequence. In some embodiments, the scale step size can be set to p, and the trackable scale size range is [p*s1, p*s n ], for example, for a scale set S = {0.6, 0.7, 0.8, 0.9, 1, 1.1, 1.2, 1.3, 1.4}, 0.6 means reducing the scale of the target to be tracked to 60% of the original, 1 means the scale of the target to be tracked remains unchanged, and if the scale step p is 0.5, the trackable scale range is [0.3, 0.7]. Therefore, the trackable scale range can be adjusted by changing the scale step without resetting the scale set. The angle set can be expressed as R = {r1, r2, …, r m}, m is the size of the angle set, r1, r2, ..., r m They represent different angle parameters, which are used to measure the degree of angle change of the target to be tracked in the video sequence. In some embodiments, the angle step size can be set to q, and the trackable angle range is [q*r1,q*r m ], for example, for the angle set R = {-2, -1, 0, 1, 2}, positive and negative represent two relative directions of angle change (for example, clockwise rotation and counterclockwise rotation, etc.), assuming that the angle step q is 5°, the trackable angle range is [-10°, 10°]. For example, -10° means rotating the angle of the target to be tracked by 10° counterclockwise, 0 means the angle remains unchanged, and 10° means rotating the angle of the target to be tracked by 10° clockwise. In addition, the trackable angle range can be adjusted by changing the angle step. In addition, the specific values of n and m can be determined according to actual conditions (for example, the content type of the video sequence, etc.). The present disclosure does not limit the size of the scale set and the angle set.

[0068] The feature extraction unit 402 can extract features of the current video frame based on one or more scale-angle parameter combinations determined by the acquired scale set and angle set, and obtain one or more feature maps under different scale-angle parameter combinations. In one embodiment, the feature is a gradient histogram (HOG) feature of the current video frame. Any method in the relevant technology can be used to extract the HOG feature, and the present disclosure is not limited to this. For example, the current video frame can be divided into a grid of size c*c (for example, 8*8), and a gradient histogram feature is calculated for each grid. The gradient histogram features of each grid are connected in series to obtain the gradient histogram feature of the current video frame.

[0069] According to an exemplary embodiment of the present disclosure, the feature extraction unit 402 may adjust the scale angle of the target to be tracked in the video frame preceding the current video frame according to each of the one or more scale angle parameter combinations, thereby obtaining the adjusted scale angles of the one or more targets to be tracked in the current video frame, and perform feature extraction on the current video frame including the targets to be tracked with the adjusted scale angles, thereby obtaining one or more feature maps. Here, the scale and angle of the target to be tracked are adjusted. Specifically, the feature extraction unit 402 may select a scale parameter and an angle parameter from the scale size range pointed to by the scale set and the angle size range pointed to by the angle set, respectively, as a scale angle parameter combination, transform the scale angle of the target to be tracked in the video frame preceding the current video frame according to the scale angle parameter combination, and then perform feature extraction on the current video frame including the target to be tracked with the transformed scale and angle, thereby obtaining one or more feature maps under the scale angle parameter combination. In this way, multiple feature maps of the current video frame may be obtained according to different angle parameter combinations. For example, if the scale set and the angle set include 3 scale parameters and 3 angle parameters respectively, there are 9 combinations of scale and angle parameters, and 9 HOG features related to the current video frame can be calculated.

[0070] The scale and angle acquisition unit 403 may obtain the scale and angle of the target to be tracked in the current video frame based on the obtained one or more feature maps, feature maps of a preset number of video frames before the current video frame, a scale set, and an angle set.

[0071] According to an exemplary embodiment of the present disclosure, the scale and angle acquisition unit 403 may obtain a first response based on feature maps (for example, gradient histogram HOG), scale sets, and angle sets of a preset number of video frames before the current video frame, and obtain one or more second responses based on the obtained one or more feature maps of the current video frame and the first response. Here, the second response reflects the degree of closeness between the adjusted scale angle of the target to be tracked in the current video frame and the true scale angle. The scale and angle acquisition unit 403 may determine the adjusted scale angle corresponding to the second response that meets the preset conditions in the one or more second responses as the scale and angle of the target in the current video frame.

[0072] Specifically, the scale and angle acquisition unit 403 can obtain the scale Gaussian response and the angle Gaussian response according to the number of scale parameters in the scale set (for example, the size n of the aforementioned scale set) and the number of angle parameters in the angle set (for example, the size m of the aforementioned angle set), respectively. Here, the scale Gaussian response, for example, but not limited to, can be expressed as G1 = {g1, g2, ..., g n},in, δ1 is a Gaussian parameter greater than 0, e is a natural constant, 1≤i≤n; the angular Gaussian response, for example, but not limited to, can be expressed as G2={h1,h2,…,h m},in, δ2 is a Gaussian parameter greater than 0, 1≤j≤m, where δ1 and δ2 can be set according to the actual target tracking scenario, and the present disclosure does not limit this. In addition, the scale and angle acquisition unit 403 can obtain feature maps of the two video frames preceding the current video frame. In some embodiments, the feature maps of the two video frames preceding the current video frame include a first feature map and a second feature map of the first video frame before the current video frame and a third feature map of the second video frame before the current video frame. Here, the difference between the first feature map and the second feature map is that in the process of obtaining the first feature map and the second feature map, the scale and angle of the target to be tracked may be different. Specifically, for the clarity of description, assuming that the first video frame before the current video frame is video frame A, and the second video frame before the current video frame is video frame B, then in the process of obtaining the first feature map, the scale and angle of the target to be tracked in video frame A are the same as the scale and angle in video frame B, respectively; in the process of obtaining the second feature map, the scale and angle of the target to be tracked in video frame A are the actual scale and actual angle in video frame A; in the process of obtaining the third feature map, the scale and angle of the target to be tracked in video frame B are the actual scale and actual angle in video frame B.

[0073] The scale and angle acquisition unit 403 may acquire a first response based on the scale Gaussian response, the angle Gaussian response, and the feature maps of the two previous video frames of the current video frame. Specifically, the scale and angle acquisition unit 403 may perform a fast Fourier transform on the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map, and the third feature map to obtain the scale Gaussian response, the angle Gaussian response, the first feature map, the second feature map, and the third feature map in the frequency domain. Then, the scale Gaussian response in the frequency domain and the angle Gaussian response in the frequency domain may be associated (for example, by multiplying the scale Gaussian response in the frequency domain and the angle Gaussian response in the frequency domain) to obtain a scale-angle integrated Gaussian response in the frequency domain. The first response is acquired based on the first feature map, the second feature map, the third feature map in the frequency domain, and the scale-angle integrated Gaussian response in the frequency domain. In some embodiments, the scale and angle acquisition unit 403 may associate the second feature map of the frequency domain and the third feature map of the frequency domain with the scale-angle integrated Gaussian response of the frequency domain (for example, multiplying the second feature map of the frequency domain and the third feature map of the frequency domain with the scale-angle integrated Gaussian response of the frequency domain, etc.) to obtain the target second feature map of the frequency domain and the target third feature map of the frequency domain, where the target second feature map of the frequency domain and the target third feature map of the frequency domain respectively carry scale-angle information (that is, carry scale information related to the aforementioned scale set and angle information related to the aforementioned angle set). The first response can be obtained based on the target second feature map of the frequency domain and the target third feature map of the frequency domain and the first feature map of the frequency domain. In one embodiment, the scale and angle acquisition unit 403 may first superimpose different weights on the target second feature map of the frequency domain and the target third feature map of the frequency domain, respectively, add the target second feature map of the frequency domain and the third feature map of the frequency domain superimposed with different weights, and then divide them by the first feature map of the frequency domain to obtain the first response.

[0074] According to an exemplary embodiment of the present disclosure, the scale and angle acquisition unit 403 may perform a fast Fourier transform on one or more feature maps related to the current video frame to obtain one or more feature maps in the frequency domain, and obtain one or more second responses based on the one or more feature maps in the frequency domain and the first response (for example, multiplying each feature map by the first response).

[0075] According to an exemplary embodiment of the present disclosure, the target tracking device 400 may further include a first response updating unit 404 ( Figure 4 (not shown in the figure), the first response updating unit 404 may update the first response according to the feature map of the current video frame and the feature map of the previous video frame of the current video frame after determining the adjusted scale angle corresponding to the second response that meets the preset conditions in one or more second responses as the scale and angle of the target in the current video frame, so as to determine the scale and size of the target to be tracked in the next video frame.

[0076] Figure 5 is a block diagram of an electronic device 500 according to an exemplary embodiment of the present disclosure.

[0077] Reference Figure 5 The electronic device 500 includes at least one memory 501 and at least one processor 502, wherein the at least one memory 501 stores a set of computer-executable instructions. When the computer-executable instruction set is executed by the at least one processor 502, the target tracking method according to the exemplary embodiment of the present disclosure is executed.

[0078] As an example, the electronic device 500 may be a PC, a tablet device, a personal digital assistant, a smart phone, or other device capable of executing the above-mentioned instruction set. Here, the electronic device 500 is not necessarily a single electronic device, but may also be any device or circuit that can execute the above-mentioned instructions (or instruction set) individually or in combination. The electronic device 500 may also be part of an integrated control system or system manager, or may be configured as a portable electronic device that is interconnected with a local or remote (e.g., via wireless transmission) interface.

[0079] In electronic device 500, processor 502 may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. By way of example and not limitation, the processor may also include an analog processor, a digital processor, a microprocessor, a multi-core processor, a processor array, a network processor, etc.

[0080] The processor 502 can execute instructions or codes stored in the memory 501, wherein the memory 501 can also store data. Instructions and data can also be sent and received over a network via a network interface device, wherein the network interface device can use any known transmission protocol.

[0081] The memory 501 may be integrated with the processor 502, for example, by placing RAM or flash memory within an integrated circuit microprocessor or the like. Furthermore, the memory 501 may comprise a separate device, such as an external disk drive, a storage array, or any other storage device usable by a database system. The memory 501 and the processor 502 may be operatively coupled or may communicate with each other, for example, via an I / O port, a network connection, or the like, such that the processor 502 can access files stored in the memory.

[0082] In addition, the electronic device 500 may further include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, a mouse, a touch input device, etc.) All components of the electronic device 500 may be connected to each other via a bus and / or a network.

[0083] According to an exemplary embodiment of the present disclosure, a computer-readable storage medium storing instructions may also be provided, wherein when the instructions are executed by at least one processor, the at least one processor is prompted to perform the target tracking method according to the present disclosure. Examples of computer-readable storage media here include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disk storage, hard disk drive (HDD), solid state drive (SSD), card storage (such as, multimedia card, secure digital (SD) card or ultra-fast digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk and any other device, any other device configured to store the computer program and any associated data, data files and data structures in a non-transitory manner and provide the computer program and any associated data, data files and data structures to a processor or computer so that the processor or computer can execute the computer program. The computer program in the above-mentioned computer-readable storage medium can be run in an environment deployed in a computer device such as a client, a host, an agent device, a server, etc. In addition, in one example, the computer program and any associated data, data files and data structures are distributed on a networked computer system so that the computer program and any associated data, data files and data structures are stored, accessed and executed in a distributed manner by one or more processors or computers.

[0084] According to an exemplary embodiment of the present disclosure, a computer program product may also be provided. Instructions in the computer program product may be executed by a processor of a computer device to implement the target tracking method according to an exemplary embodiment of the present disclosure.

[0085] According to the target tracking method, device, electronic device and storage medium disclosed herein, feature extraction is performed on the current video frame based on the scale set and angle set of the target to be tracked in the video sequence, and one or more feature maps of the current video frame under different scale-angle parameter combinations can be obtained. Based on the obtained one or more feature maps, the feature map, scale set and angle set of the video frame before the current video frame, the scale and angle of the target in the current video frame can be tracked, thereby improving the accuracy of video target tracking.

[0086] In addition, according to the target tracking method, device, electronic device and storage medium disclosed in the present invention, any video target displacement tracking method can be easily superimposed to achieve four-dimensional tracking of the target (i.e., two-dimensional displacement, one-dimensional scale size and one-dimensional angle posture).

[0087] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0088] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A target tracking method, characterized in that: include: Acquire a scale set and an angle set of a target to be tracked in a video sequence, wherein the scale set includes one or more scale parameters for measuring a degree of scale change of the target to be tracked, and the angle set includes one or more angle parameters for measuring a degree of angular change of the target to be tracked; adjusting the scale angle of the target to be tracked in a video frame previous to the current video frame according to each of the one or more scale angle parameter combinations determined by the scale set and the angle set, to obtain one or more adjusted scale angles of the target to be tracked in the current video frame; Performing feature extraction on a current video frame including the target to be tracked with the scale angle adjusted, to obtain one or more feature maps of the current video frame; Obtaining a first response based on feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set; Obtain one or more second responses according to the one or more feature maps and the first response, each second response reflecting a degree of closeness between the adjusted scale angle of the target to be tracked in the current video frame and a true scale angle; The adjusted scale and angle corresponding to the second response that meets the preset condition among the one or more second responses are determined as the scale and angle of the target to be tracked in the current video frame.

2. The target tracking method according to claim 1, wherein: The obtaining a first response according to the feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set includes: Obtaining a scale Gaussian response and an angle Gaussian response according to the number of scale parameters in the scale set and the number of angle parameters in the angle set, respectively; The first response is obtained based on the scale Gaussian response, the angle Gaussian response, and feature maps of two video frames before the current video frame.

3. The target tracking method according to claim 2, wherein: The feature maps of the first two video frames of the current video frame include a first feature map and a second feature map of a first video frame before the current video frame and a third feature map of a second video frame before the current video frame; The acquiring the first response based on the scale Gaussian response, the angle Gaussian response, and the feature maps of the two video frames before the current video frame includes: Performing a fast Fourier transform on the scale Gaussian response, the angular Gaussian response, the first feature map, the second feature map, and the third feature map to obtain a scale Gaussian response, an angular Gaussian response, a first feature map, a second feature map, and a third feature map in the frequency domain; Correlating the scale Gaussian response in the frequency domain with the angular Gaussian response in the frequency domain to obtain a scale-angle integrated Gaussian response in the frequency domain; The first response is obtained based on the first characteristic map, the second characteristic map, the third characteristic map and the scale-angle integrated Gaussian response in the frequency domain.

4. The target tracking method according to claim 3, wherein: The obtaining of the first response based on the first characteristic map, the second characteristic map, the third characteristic map and the scale angle integrated Gaussian response in the frequency domain includes: Associating the second feature map in the frequency domain and the third feature map in the frequency domain with the scale-angle integrated Gaussian response in the frequency domain, respectively, to obtain a target second feature map in the frequency domain and a target third feature map in the frequency domain, wherein the target second feature map in the frequency domain and the target third feature map in the frequency domain respectively carry scale-angle information; The first response is obtained according to the target second characteristic map in the frequency domain, the target third characteristic map in the frequency domain, and the first characteristic map in the frequency domain.

5. The target tracking method according to claim 1, wherein: The acquiring one or more second responses according to the one or more characteristic graphs and the first response includes: Performing a fast Fourier transform on the one or more feature maps to obtain one or more feature maps in the frequency domain; The one or more second responses are obtained according to each feature map in the one or more feature maps in the frequency domain and the first response.

6. The target tracking method according to claim 1, wherein: After determining the adjusted scale and angle corresponding to the second response that meets the preset condition among the one or more second responses as the scale and angle of the target to be tracked in the current video frame, the method further includes: The first response is updated according to the feature map of the current video frame and the feature map of a video frame before the current video frame.

7. A target tracking device, characterized in that: include: a set acquisition unit configured to: acquire a scale set and an angle set of a target to be tracked in a video sequence, wherein the scale set includes one or more scale parameters for measuring a degree of scale change of the target to be tracked, and the angle set includes one or more angle parameters for measuring a degree of angular change of the target to be tracked; a feature extraction unit configured to: adjust the scale angle of the target to be tracked in a video frame previous to the current video frame according to each of the one or more scale angle parameter combinations determined by the scale set and the angle set, to obtain one or more adjusted scale angles of the target to be tracked in the current video frame; and perform feature extraction on the current video frame including the target to be tracked with the adjusted scale angles, to obtain one or more feature maps of the current video frame; The scale and angle acquisition unit is configured to: acquire a first response based on feature maps of a preset number of video frames before the current video frame, the scale set, and the angle set; acquire one or more second responses based on the one or more feature maps and the first responses, each second response reflecting the degree of proximity between the adjusted scale angle of the target to be tracked in the current video frame and the true scale angle; The adjusted scale and angle corresponding to the second response that meets the preset condition among the one or more second responses are determined as the scale and angle of the target to be tracked in the current video frame.

8. The target tracking device according to claim 7, wherein: The scale and angle acquisition unit is configured to: Obtaining a scale Gaussian response and an angle Gaussian response according to the number of scale parameters in the scale set and the number of angle parameters in the angle set, respectively; The first response is obtained based on the scale Gaussian response, the angle Gaussian response, and feature maps of two video frames before the current video frame.

9. The target tracking device according to claim 8, wherein: The feature maps of the first two video frames of the current video frame include a first feature map and a second feature map of a first video frame before the current video frame and a third feature map of a second video frame before the current video frame; The scale and angle acquisition unit is configured to: Performing a fast Fourier transform on the scale Gaussian response, the angular Gaussian response, the first feature map, the second feature map, and the third feature map to obtain a scale Gaussian response, an angular Gaussian response, a first feature map, a second feature map, and a third feature map in the frequency domain; Correlating the scale Gaussian response in the frequency domain with the angular Gaussian response in the frequency domain to obtain a scale-angle integrated Gaussian response in the frequency domain; The first response is obtained based on the first characteristic map, the second characteristic map, the third characteristic map and the scale-angle integrated Gaussian response in the frequency domain.

10. The target tracking device according to claim 9, wherein: The scale and angle acquisition unit is configured to: Associating the second feature map in the frequency domain and the third feature map in the frequency domain with the scale-angle integrated Gaussian response in the frequency domain, respectively, to obtain a target second feature map in the frequency domain and a target third feature map in the frequency domain, wherein the target second feature map in the frequency domain and the target third feature map in the frequency domain respectively carry scale-angle information; The first response is obtained according to the target second characteristic map in the frequency domain, the target third characteristic map in the frequency domain, and the first characteristic map in the frequency domain.

11. The target tracking device according to claim 7, wherein: The scale and angle acquisition unit is configured to: Performing a fast Fourier transform on the one or more feature maps to obtain one or more feature maps in the frequency domain; The one or more second responses are obtained according to each feature map in the one or more feature maps in the frequency domain and the first response.

12. The target tracking device according to claim 7, wherein: The first response updating unit is configured to: After determining the adjusted scale angle corresponding to the second response that meets the preset conditions in the one or more second responses as the scale and angle of the target to be tracked in the current video frame, the first response is updated according to the feature map of the current video frame and the feature map of the previous video frame of the current video frame.

13. An electronic device, characterized in that: include: at least one processor; at least one memory storing computer-executable instructions, When the computer executable instructions are executed by the at least one processor, the computer executable instructions prompt the at least one processor to execute the target tracking method according to any one of claims 1 to 6.

14. A computer-readable storage medium storing instructions, characterized in that: When the instructions are executed by at least one processor, the at least one processor is prompted to perform the target tracking method according to any one of claims 1 to 6.

15. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by at least one processor, the target tracking method according to any one of claims 1 to 6 is implemented.

Images