A three-dimensional model-based motion trajectory generation method, device and server

Abstract

The application is suitable for the technical field of image processing, and provides a motion trajectory generation method based on a three-dimensional model, which comprises the following steps: acquiring video data; obtaining a target motion trajectory corresponding to a target object according to the video data; acquiring a scene image according to the target motion trajectory; generating a three-dimensional model according to the scene image; adding the target motion trajectory to the three-dimensional model; and displaying the three-dimensional model after the target motion trajectory is added to the user. It can be seen that the application can achieve the effect of intuitively and high-precision monitoring the target motion trajectory of the target object.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of image processing, and particularly relates to a motion trajectory generation method and device based on a three-dimensional model and a server. BACKGROUND

[0002] With the development of intelligent video monitoring, a large amount of information of moving objects can be observed by video monitoring. However, for some special scenes, it is difficult to meet the needs of users to intuitively and accurately monitor the motion trajectory of a target object. SUMMARY

[0003] The embodiments of the present application provide a motion trajectory generation method and device based on a three-dimensional model and a server, which can solve the technical problem that the prior art cannot meet the needs of users to intuitively and accurately detect the motion trajectory of a target object.

[0004] In a first aspect, the embodiments of the present application provide a motion trajectory generation method based on a three-dimensional model, comprising:

[0005] obtaining video data;

[0006] obtaining a target motion trajectory corresponding to a target object according to the video data;

[0007] obtaining a scene image according to the target motion trajectory;

[0008] generating a three-dimensional model according to the scene image;

[0009] adding the target motion trajectory to the three-dimensional model;

[0010] displaying the three-dimensional model after adding the target motion trajectory to a user.

[0011] In a possible implementation manner of the first aspect, the target motion trajectory corresponding to the target object is obtained according to the video data, comprising:

[0012] determining a target object in the video data;

[0013] generating a target motion trajectory corresponding to the target object.

[0014] In a possible implementation manner of the first aspect, the target object in the video data is determined, comprising:

[0015] inputting the video data into a target detection model to output a candidate object;

[0016] identifying identity information of the candidate object;

[0017] determining a target object in the candidate object according to the identity information.

[0018] In a possible implementation manner of the first aspect, the target motion trajectory corresponding to the target object is generated, including:

[0019] An initial motion trajectory of the target object is acquired;

[0020] The initial motion trajectory is preprocessed;

[0021] Anomaly detection is performed on the preprocessed initial motion trajectory to obtain a target motion trajectory.

[0022] In a possible implementation manner of the first aspect, the initial motion trajectory of the target object is acquired, including:

[0023] A position and a target feature corresponding to the target object in the video data are marked;

[0024] The marked video data is input into a target tracking model to output an initial motion trajectory corresponding to the target object.

[0025] In a possible implementation manner of the first aspect, the initial motion trajectory is preprocessed, including:

[0026] The initial motion trajectory is compressed;

[0027] Similarity measurement is performed on the compressed initial motion trajectory;

[0028] The initial motion trajectory after the similarity measurement is clustered.

[0029] In the first aspect, an embodiment of the present application provides a motion trajectory generation device based on a three-dimensional model, including:

[0030] A first acquisition module is configured to acquire video data;

[0031] A first generation module is configured to obtain a target motion trajectory corresponding to a target object according to the video data;

[0032] A second acquisition module is configured to acquire a scene image according to the target motion trajectory;

[0033] A second generation module is configured to generate a three-dimensional model according to the scene image;

[0034] An adding module is configured to add the target motion trajectory to the three-dimensional model;

[0035] A display module is configured to display the three-dimensional model after the target motion trajectory is added to the user.

[0036] In a possible implementation manner of the second aspect, the first generation module includes:

[0037] determining a target object in the video data;

[0038] generating a target motion trajectory corresponding to the target object

[0039] In a possible implementation manner of the second aspect, the determining sub-module comprises:

[0040] a detection unit configured to input the video data into a target detection model and output a candidate object;

[0041] an identification unit configured to identify identity information of the candidate object;

[0042] a determination unit configured to determine a target object in the candidate object according to the identity information.

[0043] In a possible implementation manner of the second aspect, the generating sub-module comprises:

[0044] an acquisition unit configured to acquire an initial motion trajectory of a target object;

[0045] a preprocessing unit configured to preprocess the initial motion trajectory;

[0046] an anomaly detection unit configured to perform anomaly detection on the preprocessed initial motion trajectory to obtain a target motion trajectory

[0047] In a possible implementation manner of the second aspect, the acquisition unit comprises:

[0048] a marking sub-unit configured to mark a position and a target feature corresponding to a target object in the video data;

[0049] a tracking sub-unit configured to input the marked video data into a target tracking model and output an initial motion trajectory corresponding to the target object.

[0050] In a possible implementation manner of the second aspect, the preprocessing unit comprises:

[0051] a compression sub-unit configured to compress the initial motion trajectory;

[0052] a similarity measurement unit configured to perform similarity measurement on the compressed initial motion trajectory;

[0053] a clustering unit configured to perform clustering on the initial motion trajectory after similarity measurement.

[0054] In a possible implementation manner, the second generation module comprises:

[0055] an extraction sub-module configured to extract feature points of the scene image;

[0056] The generating sub-module is configured to generate a point cloud according to the depth information and the color information of the feature points.

[0057] The three-dimensional reconstruction sub-module is configured to perform three-dimensional reconstruction based on the point cloud to obtain the three-dimensional model.

[0058] In a third aspect, an embodiment of the present application provides a server, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method in the first aspect when executing the computer program.

[0059] In a fourth aspect, an embodiment of the present application provides a readable storage medium, and the computer program is executable on the processor to implement the method in the first aspect.

[0060] Compared with the prior art, the embodiment of the present application has the beneficial effects that:

[0061] In the embodiment of the present application, the video data is acquired, the target motion trajectory corresponding to the target object is obtained according to the video data, the scene image is acquired according to the target motion trajectory, the three-dimensional model is generated according to the scene image, the target motion trajectory is added to the three-dimensional model, and the three-dimensional model after adding the target motion trajectory is displayed to the user. It can be seen that the embodiment of the present application can achieve the effect of intuitively and accurately monitoring the motion trajectory of the target object. BRIEF DESCRIPTION OF DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0063] Figure 1 is a flow diagram of the motion trajectory generation method based on the three-dimensional model provided by the embodiment of the present application;

[0064] Figure 2 is a structural block diagram of the motion trajectory generation device based on the three-dimensional model provided by the embodiment of the present application;

[0065] Figure 3 is a structural diagram of the server provided by the embodiment of the present application. DETAILED DESCRIPTION

[0066] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, technologies, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

[0067] It is to be understood that the terminology "includes", "has", "holds", "contains" and / or "comprising", "including", "having" and their conjugates, as used herein, means "including but not limited to", and not to the exclusion of any other term or aspect.

[0068] It is also to be understood that the terminology "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items, as well as to the possibility that one or more of the listed items can be excluded from the combination.

[0069] As used in the description of the application and the appended claims, the term "if' can be interpreted to mean "when" or "upon" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrase "if it is determined" or "if [a described condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon [the described condition or event] being detected" or "in response to [the described condition or event] being detected", depending on the context.

[0070] In addition, the terms "first", "second", "third", etc. as used in the description of embodiments herein and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of use in either order.

[0071] The terms "comprise", "comprising", "include", "including", "contain", "containing", "have", "having", and the like are used in the detailed description and in the claims of this application in a manner to mean that the described feature, step, or component is included, but not limited to, the specific feature, step, or component. The term "comprise" is used in the sense of "including but not limited to".

[0072] The technical solutions provided by the embodiments of the present application will be described below through specific embodiments.

[0073] Referring toFigure 1 A flowchart of a three-dimensional model-based motion trajectory generation method provided by an embodiment of the present application is shown in FIG. 1. As an example but not limitation, the method can be applied to a server. The method can include the following steps:

[0074] In step S101, video data is obtained.

[0075] The video data is video monitoring data transmitted by a camera arranged in a preset area (e.g., within a range of 1 km from a crime scene).

[0076] It can be understood that in the embodiment of the present application, the video monitoring data transmitted by the camera arranged in the preset area is obtained, so as to observe the motion trajectories of a large number of objects (e.g., outdoor and indoor pedestrians). After the target object is determined, the motion trajectory of the target object is generated, and the scene image captured according to the motion trajectory is obtained. The three-dimensional model is generated according to the scene image, and finally the motion opportunity of the target object is added to the three-dimensional model, so as to intuitively and accurately monitor the motion trajectory of the target object by a user (e.g., a manager).

[0077] In step S102, a target motion trajectory corresponding to the target object is obtained according to the video data.

[0078] In a specific application, the target motion trajectory corresponding to the target object is obtained according to the video data, including:

[0079] In step S201, the target object in the video data is determined.

[0080] In a specific application, the target object in the video data is determined, including:

[0081] In step S301, the video data is input into a target detection model to output a candidate object.

[0082] As an example, the video sequence is vectorized to obtain a plurality of continuous images represented by vectors. The feature similarity of adjacent images is evaluated by using a selective search algorithm. The images with high similarity are combined into a candidate box. The candidate box is input into a CNN convolutional neural network to obtain deep features. Finally, the deep features are classified by SVM and regressed by RNN to obtain a classification result, and the candidate object is determined.

[0083] In step S302, the identity information of the candidate object is identified.

[0084] As an example, the feature points of the image area corresponding to the candidate object are extracted, and the feature points are encoded to obtain a feature vector value. The feature vector value is input into a preset face matching library to obtain the identity information of the candidate object.

[0085] In step S303, the target object in the candidate object is determined according to the identity information.

[0086] It can be understood that the target object is determined in the database according to the identity information.

[0087] In step S202, a target motion trajectory corresponding to the target object is generated.

[0088] In a specific application, generating the target motion trajectory corresponding to the target object includes:

[0089] In step S401, an initial motion trajectory of the target object is acquired.

[0090] In a specific application, acquiring the initial motion trajectory of the target object includes:

[0091] In step S501, a position corresponding to the target object and a target feature in video data are marked.

[0092] In step S502, the marked video data is input into a target tracking model, and an initial motion trajectory corresponding to the target object is output.

[0093] For example, the position corresponding to the target object and the target feature in the video data are determined by using a YOLO target detection algorithm, then the similarity between each candidate frame is calculated, a Mahalanobis distance or a cosine distance is used as a basis for target management, and finally, a Hungarian algorithm or a / KM algorithm is used to assign the target, two targets with high similarity are associated as one trajectory, and the initial motion trajectory is obtained.

[0094] In step S402, the initial motion trajectory is preprocessed.

[0095] In a specific application, preprocessing the initial motion trajectory includes:

[0096] In step S601, the initial motion trajectory is compressed.

[0097] In step S602, the similarity of the compressed initial motion trajectory is measured.

[0098] In step S603, the initial motion trajectory after similarity measurement is clustered.

[0099] For example, first, a trajectory compression algorithm (such as a static grid division method) is used to replace the original coordinate points with coordinate points that can preserve the structure of the trajectory, thereby reducing the amount of trajectory data points as much as possible under the condition of meeting the similarity of the compressed trajectory and the original trajectory; then, a similarity algorithm (such as Euclidean distance or edit distance) is used to measure the distance between trajectories to measure the similarity of the trajectories, and trajectories with high similarity are combined together; finally, a trajectory clustering algorithm (such as a K-Mens algorithm) is used to divide a large number of collected trajectories into relatively homogeneous clusters, and the preprocessed initial motion trajectory is obtained.

[0100] Step S403, the pre-processed initial motion trajectory is detected for abnormality to obtain a target motion trajectory.

[0101] Exemplarily, the pre-processed data is divided into a training set and a test set according to a certain proportion, and a clustering algorithm is selected for learning to establish a normal motion trajectory mode; whether the motion of the current object is normal is judged by the normal motion trajectory mode obtained by training to obtain a target motion trajectory.

[0102] Step S103, a scene image corresponding to a scene passed by the target object is retrieved from a local database according to the target motion trajectory.

[0103] It can be understood that the scene image corresponding to the scene passed by the target object is retrieved from the local database according to the target motion trajectory, and the scene image can be obtained by a depth camera.

[0104] Step S104, a three-dimensional model is generated according to the scene image.

[0105] In a specific application, the three-dimensional model is generated according to the scene image, including:

[0106] Step S701, a feature point of the scene image is extracted.

[0107] Exemplarily, the feature point of the scene image is extracted according to a preset feature extraction algorithm, wherein the preset feature extraction algorithm can be a corner detection algorithm, such as Harris corner detection, FAST corner detection, or a patch feature point detection algorithm, such as SIFI extraction algorithm, SURF extraction algorithm.

[0108] Step S702, a point cloud is generated according to depth information and color information of the feature point.

[0109] Exemplarily, the feature point is directly processed according to an SFM algorithm, the depth information corresponding to the feature point is calculated, and the color information corresponding to the feature point is extracted, so that the feature point with the depth information and the color information is regarded as a point cloud.

[0110] Step S703, a three-dimensional reconstruction is performed based on the point cloud to obtain a three-dimensional model.

[0111] Exemplarily, a three-dimensional reconstruction algorithm (such as MVE algorithm) is adopted to extract a surface triangle of the point cloud to generate a three-dimensional model.

[0112] Step S105, the target motion trajectory is added to the three-dimensional model.

[0113] Exemplarily, the target motion trajectory is repositioned to map the target motion trajectory from a coordinate system to the three-dimensional model.

[0114] Step S106, display the three-dimensional model after adding the target motion track to the user.

[0115] Illustratively, the user sends a scene ID to the server through the user terminal, and then the server sends the three-dimensional model added with the target motion track to the user terminal to display the three-dimensional model added with the target motion track to the user.

[0116] In the embodiments of the present application, the video data is obtained, the target motion track corresponding to the target object is obtained according to the video data, the scene image is obtained according to the target motion track, the three-dimensional model is generated according to the scene image, the target motion track is added to the three-dimensional model, and the three-dimensional model after adding the target motion track is displayed to the user. It can be seen that the embodiments of the present application can achieve the effect of intuitively and accurately monitoring the motion track of the target object.

[0117] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0118] Corresponding to the method described in the above embodiments, Figure 2 The structure block diagram of the motion track generation device based on the three-dimensional model provided by the embodiments of the present application is shown, and only the part related to the embodiments of the present application is shown for the convenience of description.

[0119] Referring to Figure 2 The device comprises:

[0120] The first obtaining module 21 is configured to obtain video data.

[0121] The first generation module 22 is configured to obtain the target motion track corresponding to the target object according to the video data.

[0122] The second obtaining module 23 is configured to obtain the scene image according to the target motion track.

[0123] The second generation module 24 is configured to generate the three-dimensional model according to the scene image.

[0124] The adding module 25 is configured to add the target motion track to the three-dimensional model.

[0125] The display module 26 is configured to display the three-dimensional model after adding the target motion track to the user.

[0126] In a possible implementation, the first generation module comprises:

[0127] The determination sub-module is configured to determine the target object in the video data.

[0128] The generating sub-module is configured to generate a target motion trajectory corresponding to the target object

[0129] In a possible implementation, the determining sub-module comprises:

[0130] The detection unit is configured to input the video data into a target detection model and output a candidate object;

[0131] The recognition unit is configured to recognize identity information of the candidate object;

[0132] The determining unit is configured to determine a target object from the candidate object according to the identity information.

[0133] In a possible implementation, the generating sub-module comprises:

[0134] The acquisition unit is configured to acquire an initial motion trajectory of a target object;

[0135] The preprocessing unit is configured to preprocess the initial motion trajectory;

[0136] The anomaly detection unit is configured to perform anomaly detection on the preprocessed initial motion trajectory to obtain a target motion trajectory

[0137] In a possible implementation, the acquisition unit comprises:

[0138] The marking sub-unit is configured to mark a position and a target feature corresponding to a target object in the video data;

[0139] The tracking sub-unit is configured to input the marked video data into a target tracking model and output an initial motion trajectory corresponding to the target object.

[0140] In a possible implementation, the preprocessing unit comprises:

[0141] The compression sub-unit is configured to compress the initial motion trajectory;

[0142] The similarity measurement unit is configured to perform similarity measurement on the compressed initial motion trajectory;

[0143] The clustering unit is configured to perform clustering on the initial motion trajectory after similarity measurement.

[0144] In a possible implementation, the second generating module comprises:

[0145] The extraction sub-module is configured to extract feature points of the scene image;

[0146] The generating sub-module is configured to generate a point cloud according to depth information and color information of the feature points;

[0147] A three-dimensional reconstruction submodule is configured to perform three-dimensional reconstruction based on the point cloud to obtain the three-dimensional model.

[0148] It should be noted that the information interaction and execution process between the above apparatuses / units are based on the same concept as the method embodiments of the present application, and the specific functions and technical effects brought by the same can be referred to the method embodiments part. Therefore, no further description is given here.

[0149] Figure 3 A structural diagram of a server provided by the embodiments of the present application is shown in FIG. 3. As shown in FIG. 3, the server 3 of the embodiment includes at least one processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30. The processor 30 implements the steps in any of the above method embodiments when executing the computer program 32. Figure 3

[0150] The server 3 can be a cloud server or other computing device. The server can include, but is not limited to, the processor 30 and the memory 31. Those skilled in the art can understand that the server can include more or fewer components, or combine certain components, or include different components, such as an input / output device, a network access device, etc. Figure 3 The server 3 shown in FIG. 3 is merely an example and does not limit the server 3. The server 3 can include more or fewer components than those shown in the figure, or combine certain components, or include different components, such as an input / output device, a network access device, etc.

[0151] The processor 30 can be a central processing unit (CPU). The processor 30 can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0152] ​The memory 31 can be an internal storage unit of the server 3, such as a hard disk or a memory of the server 3, in some embodiments. The memory 31 can also be an external storage device of the server 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the server 3, in other embodiments. Further, the memory 31 can include both the internal storage unit and the external storage device of the server 3. The memory 31 is used to store an operating system, an application program, a boot loader, data, and other programs, etc., such as program codes of the computer program, etc. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0153] It can be clearly understood by those skilled in the art that, for the convenience and brevity of description, only the division of the above functional units and modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to complete all or part of the above-described functions. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit, and the integrated unit can be realized in the form of hardware or in the form of a software functional unit. In addition, the specific names of the functional units and modules are only for the convenience of mutual distinction, and do not limit the protection scope of the present application. The specific working process of the units and modules in the system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0154] The embodiments of the present application further provide a readable storage medium, which is preferably a computer readable storage medium, and stores a computer program. The computer program is executed by a processor to realize the steps in each of the above method embodiments.

[0155] The embodiments of the present application provide a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal is caused to execute the steps in each of the above method embodiments.

[0156] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the present application can implement all or part of the processes in the above-mentioned embodiment methods through a computer program to instruct relevant hardware to complete, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer readable medium at least includes any entity or device capable of carrying the computer program code to the server, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, magnetic disk or optical disk, etc. In some jurisdictions, according to legislation and patent practice, the computer readable medium can not be an electrical carrier signal and a telecommunication signal.

[0157] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0158] Those skilled in the art can appreciate that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0159] In the embodiments provided in the present application, it should be understood that the disclosed server and method can be implemented in other ways. For example, the above-described device / network equipment embodiment is only schematic, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed each other can be indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0160] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may also be distributed to multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiment scheme according to actual needs.

[0161] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for generating motion trajectories based on a three-dimensional model, characterized in that, include: Acquire video data; Obtaining the target motion trajectory corresponding to the target object based on the video data includes: determining the target object in the video data, including: inputting the video data into a target detection model and outputting candidate objects; identifying the identity information of the candidate objects; determining the target object among the candidate objects based on the identity information; vectorizing the video sequence to obtain multiple continuous images with vectorized representations, evaluating the feature similarity of adjacent images using a selective search algorithm, merging images with high similarity into the same candidate box, inputting the candidate box into a CNN convolutional neural network to obtain depth features, and finally performing SVM classification and RNN regression on the depth features to obtain classification results and determine the candidate objects; generating the target motion trajectory corresponding to the target object, including: obtaining the initial motion trajectory of the target object, including: marking the position and target features corresponding to the target object in the video data; inputting the marked video data into a target tracking model and outputting the initial motion trajectory corresponding to the target object; and using the YOLO target detection algorithm to determine the target object's position and target features based on the position and target features in the video data. The system first identifies the location and target features, then calculates the similarity between candidate boxes in adjacent frames, using Mahalanobis distance or cosine distance similarity algorithms as the basis for target management. Finally, it assigns targets using the Hungarian algorithm or the KM algorithm, associating two highly similar targets with a single trajectory to obtain the initial motion trajectory. The initial motion trajectory is then preprocessed, including: compressing the initial motion trajectory; measuring the similarity of the compressed initial motion trajectory; clustering the similarity-measured initial motion trajectories. First, a trajectory compression algorithm, while satisfying the similarity requirement between the compressed trajectory and the original trajectory, replaces the original coordinate points with coordinate points that retain the trajectory structure, reducing the amount of trajectory data points. Next, a similarity algorithm is used to measure the similarity between trajectories using the distance between them, merging highly similar trajectories together. Finally, a trajectory clustering algorithm is used to divide the large number of collected trajectories into relatively homogeneous clusters, obtaining the preprocessed initial motion trajectory. Anomaly detection is then performed on the preprocessed initial motion trajectory to obtain the target motion trajectory. Obtain a scene image based on the target's motion trajectory; Generate a 3D model based on the scene image; Add the target motion trajectory to the 3D model; The user is shown a 3D model of the target motion trajectory after it has been added.

2. The motion trajectory generation method based on a three-dimensional model as described in claim 1, characterized in that, Generating a 3D model based on the scene image includes: Extract feature points from the scene image; Generate a point cloud based on the depth and color information of the feature points; The three-dimensional model is obtained by performing three-dimensional reconstruction based on the point cloud.

3. A motion trajectory generation device based on a three-dimensional model, characterized in that, include: The first acquisition module is used to acquire video data; The first generation module is used to obtain the target motion trajectory corresponding to the target object based on the video data, including: determining the target object in the video data, including: inputting the video data into a target detection model and outputting candidate objects; identifying the identity information of the candidate objects; determining the target object among the candidate objects based on the identity information; vectorizing the video sequence to obtain multiple continuous images with vectorized representations, evaluating the feature similarity of adjacent images using a selective search algorithm, merging images with high similarity into the same candidate box, inputting the candidate box into a CNN convolutional neural network to obtain depth features, and finally performing SVM classification and RNN regression on the depth features to obtain classification results and determine the candidate objects; generating the target motion trajectory corresponding to the target object, including: obtaining the initial motion trajectory of the target object, including: marking the position and target features corresponding to the target object in the video data; inputting the marked video data into a target tracking model and outputting the initial motion trajectory corresponding to the target object; and using the YOLO target detection algorithm to determine the target object in the video data. The system first identifies the location and target features of the object, then calculates the similarity between candidate boxes in adjacent frames, using Mahalanobis distance or cosine distance similarity algorithms as the basis for target management. Finally, it assigns targets using the Hungarian algorithm or the KM algorithm, associating two highly similar targets with a single trajectory to obtain the initial motion trajectory. The initial motion trajectory is then preprocessed, including: compressing the initial motion trajectory; measuring the similarity of the compressed initial motion trajectory; clustering the similarity-measured initial motion trajectories. First, a trajectory compression algorithm, while satisfying the similarity requirement between the compressed trajectory and the original trajectory, replaces the original coordinate points with coordinate points that retain the trajectory structure, reducing the amount of trajectory data points. Next, a similarity algorithm is used to measure the similarity between trajectories using the distance between them, merging highly similar trajectories together. Finally, a trajectory clustering algorithm is used to divide the large number of collected trajectories into relatively homogeneous clusters, obtaining the preprocessed initial motion trajectory. Anomaly detection is then performed on the preprocessed initial motion trajectory to obtain the target motion trajectory. The second acquisition module is used to acquire scene images based on the target motion trajectory; The second generation module is used to generate a three-dimensional model based on the scene image; An add module is used to add the target motion trajectory to the 3D model; The display module is used to display the three-dimensional model of the target motion trajectory to the user.

4. A server, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 2.

5. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 2.

Images