Trigger line setting method and device and network equipment

Abstract

The invention is suitable for the technical field of image processing, and provides a trigger line setting method and device and network equipment, and the method comprises the steps: obtaining image information and depth information of a target scene; generating a three-dimensional model of the target scene according to the image information and the depth information; obtaining to-be-detected target information; based on the three-dimensional model of the target scene and the information of the to-be-detected target, generating a simulation video of the to-be-detected target in the target scene; performing iterative optimization on an initially set trigger line according to the simulation video to obtain an optimized trigger line; and setting the trigger line of the target scene according to the optimized trigger line. Through the method, the setting efficiency and accuracy of the trigger line can be improved.

Description

Technical Field

[0001] This application belongs to the technical field of image processing, and particularly relates to a method and device for setting a trigger line, a network device, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the development of image technology, people can utilize image technology for their services. For example, when a user needs to capture a vehicle passing through a specified area, a camera can be installed, and the field of view of the camera can include the specified area. Then, a trigger line can be set at the corresponding position of the specified area in the captured image. When a vehicle passes by, the captured image is analyzed through image analysis technology. If it is determined that the vehicle is at the position of the trigger line, a photographing action is triggered.

[0003] Currently, the method for setting a trigger line usually requires on-site manual setting, and the result of manual setting often depends on the experience of the operator. Therefore, the existing method has low efficiency and it is difficult to ensure the setting effect.

[0004] Therefore, a new method is needed to solve the above technical problems. Summary of the Invention

[0005] Embodiments of this application provide a method and device for setting a trigger line and a network device, which can solve the problems of low efficiency and low accuracy in setting the existing trigger line.

[0006] In a first aspect, embodiments of this application provide a method for setting a trigger line, including:

[0007] Obtaining image information and depth information of a target scene;

[0008] Generating a three-dimensional model of the target scene according to the image information and the depth information;

[0009] Obtaining information of a target to be measured;

[0010] Generating a simulation video of the target to be measured in the target scene based on the three-dimensional model of the target scene and the information of the target to be measured;

[0011] Iteratively optimizing an initially set trigger line according to the simulation video to obtain an optimized trigger line;

[0012] Setting the trigger line of the target scene according to the optimized trigger line.

[0013] In the embodiments of the present application, a three-dimensional model of the target scene is generated based on the image information and depth information of the target scene. According to the three-dimensional model and the obtained information of the target to be measured, a simulation video of the target to be measured within the target scene is generated. Since the simulation video is generated based on the image information and depth information of the target scene, the simulation video can reflect the relevant information of the target scene, so that the user can view the relevant information of the target scene from the simulation video without reaching the specific scene. In addition, since the simulation video also includes the target to be measured, when the initially set trigger line is iteratively optimized according to the generated simulation video, it is equivalent to detecting the target to be measured with different trigger lines, which is conducive to determining a better trigger line as the trigger line of the target scene. That is, when setting the trigger line by the above method, since the setting can be performed without the user arriving at the scene, the setting efficiency of the trigger line is improved. In addition, since the trigger line of the target scene is set according to the iteratively optimized trigger line, the influence of personal experience on the set trigger line is reduced, which is conducive to improving the accuracy of the set trigger line.

[0014] Optionally, the obtaining of the image information and depth information of the target scene includes:

[0015] Obtaining the image information of the target scene and the depth information corresponding to the image information through a dual-mode acquisition device;

[0016] Or,

[0017] Obtaining the image information of the target scene through an image acquisition device, and outputting the depth information corresponding to the image information of the target scene through a pre-trained depth estimation model, where the pre-trained depth estimation model is trained according to the training image information and training depth information of the target scene.

[0018] Optionally, the obtaining of the image information and depth information of the target scene includes:

[0019] Obtaining the image information and depth information of the target scene through a shooting device at different positions and / or different perspectives in the target scene;

[0020] The generating of the three-dimensional model of the target scene according to the image information and the depth information includes:

[0021] Determining the three-dimensional pose of the shooting device in the target coordinate system according to each piece of the image information and the depth information to obtain the corresponding three-dimensional pose of the shooting device, where the target coordinate system is the coordinate system where the target scene is located;

[0022] Generate a 3D model of the target scene based on the 3D poses of the respective imaging devices, the image information, and the depth information.

[0023] Optionally, generating a simulation video of the target to be measured in the target scene based on the 3D model of the target scene and the information of the target to be measured includes:

[0024] Determine the image features at a specified angle in the 3D model of the target scene to obtain background image features;

[0025] Generate a simulation video of the target to be measured in the target scene based on the background image features, the 3D model of the target scene, and the information of the target to be measured.

[0026] Optionally, the information of the target to be measured includes the text description information and the coordinate position information of the target to be measured. Generating a simulation video of the target to be measured in the target scene based on the background image features, the 3D model of the target scene, and the information of the target to be measured includes:

[0027] Extract the features of the text description information of the target to be measured to obtain text features, and extract the features of the coordinate position information to obtain digital features;

[0028] Perform alignment processing on the text features, the digital features, and the background image features to obtain the aligned text features, the aligned digital features, and the aligned background image features;

[0029] Generate a simulation video of the target to be measured in the target scene according to the aligned text features, the aligned digital features, the aligned background image features, and the 3D model of the target scene.

[0030] Optionally, the simulation video is used to reflect the movement process of the target to be measured in continuous time. Iteratively optimizing the initially set trigger line according to the simulation video to obtain an optimized trigger line includes:

[0031] During the playback of the simulation video, detect whether the target to be measured in the currently played video frame is on the initially set trigger line;

[0032] If the target to be measured in the currently played video frame is on the initially set trigger line, perform a capture process on the simulation video to obtain a captured image;

[0033] Analyze the target to be measured in the captured image to obtain an analysis result;

[0034] If the analysis result does not meet the preset requirements, adjust the initially set trigger line to obtain the adjusted trigger line;

[0035] Use the adjusted trigger line as the newly initially set trigger line, and return to the step of detecting whether the target to be measured in the currently played video frame is on the initially set trigger line during the playback of the simulation video and subsequent steps until the analysis result meets the preset requirements.

[0036] Optionally, the simulation video is used to reflect the movement process of the target to be measured within a continuous time. Iteratively optimizing the initially set trigger line according to the simulation video to obtain the optimized trigger line includes:

[0037] During the playback of the simulation video, detect whether the target to be measured in the currently played video frame is on the initially set trigger line;

[0038] If the target to be measured in the currently played video frame is on the initially set trigger line, control the barrier gate to lift or lower;

[0039] Statistically calculate the duration from the moment when it is detected that the target to be measured is on the initially set trigger line to the moment when the control of the barrier gate is completed to obtain T1;

[0040] If T1 is greater than the preset duration threshold T2, adjust the position parameter of the initially set trigger line so that its position in the image is away from the focal plane of the shooting device to obtain the adjusted trigger line, where T2 is the duration required for the target to be measured to move from the position of the trigger line to the barrier gate at the preset moving speed;

[0041] Use the adjusted trigger line as the newly initially set trigger line, and return to the step of detecting whether the target to be measured in the currently played video frame is on the initially set trigger line during the playback of the simulation video and subsequent steps until the obtained T is not greater than the preset duration threshold T2.

[0042] In a second aspect, an embodiment of the present application provides a trigger line setting device, including:

[0043] A target scene information acquisition module, configured to acquire the image information and depth information of the target scene;

[0044] A three-dimensional model generation module of the target scene, configured to generate a three-dimensional model of the target scene according to the image information and the depth information;

[0045] A target to be measured information acquisition module, configured to acquire the target to be measured information;

[0046] A simulation video generation module, configured to generate a simulation video of the target to be measured in the target scenario based on the three-dimensional model of the target scenario and the information of the target to be measured;

[0047] A trigger line optimization module, configured to iteratively optimize the initially set trigger line according to the simulation video to obtain an optimized trigger line;

[0048] An optimized trigger line setting module, configured to set the trigger line of the target scenario according to the optimized trigger line.

[0049] In a third aspect, an embodiment of the present application provides a network device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in any item of the first aspect is implemented.

[0050] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, and when the computer program is executed by a processor, the method described in any item of the first aspect is implemented.

[0051] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product runs on a network device, the network device is enabled to execute the method described in any item of the first aspect above.

[0052] It can be understood that the beneficial effects of the above second aspect to fifth aspect can refer to the relevant descriptions in the first aspect above, and will not be repeated here. Description of the Drawings

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art.

[0054] Figure 1 It is a schematic flowchart of a method for setting a trigger line provided by an embodiment of the present application;

[0055] Figure 2 It is a schematic structural diagram of a device for setting a trigger line provided by an embodiment of the present application;

[0056] Figure 3 It is a schematic structural diagram of a network device provided by an embodiment of the present application. Detailed Embodiments

[0057] In the following description, specific details such as specific system architectures and technologies are presented for purposes of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from obscuring the description of the present application.

[0058] It should be understood that when used in the specification of the present application and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0059] It should also be understood that the term "and / or" used in the specification of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0060] Reference to "one embodiment" or "some embodiments" etc. described in the specification of the present application means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the present application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways.

[0061] Currently, when it is necessary to capture vehicles passing through a certain area (such as capturing vehicles in a no-parking area), it is usually necessary to manually test on-site before setting the corresponding trigger line, which is also called a detection line, a capture line, an induction line, etc. After setting the trigger line, if it is detected that a vehicle passes through the area corresponding to the trigger line, a capture action is performed.

[0062] Since manual on-site testing is required to set the trigger line, and it takes a certain amount of time for personnel to reach the site, the efficiency of setting the trigger line is low. In addition, since the setting of the trigger line is related to personal experience, the setting effect is unstable.

[0063] In order to improve the efficiency of setting the trigger line and the stability of the setting effect, the method for setting the trigger line provided by the embodiments of the present application is described below with reference to the accompanying drawings.

[0064] Figure 1 The flowchart of a method for setting a trigger line provided by an embodiment of the present application is shown. This setting method is applied to a network device and is described in detail as follows:

[0065] S11, obtain the image information and depth information of the target scene.

[0066] Among them, the target scene is the scene where the trigger line is to be set. When the user needs to monitor a target passing through area A inside his house, a trigger line can be set in the corresponding picture area of area A. At this time, the scene corresponding to area A is the target scene. When the staff needs to monitor a vehicle staying in the no-parking area B, a trigger line can be set in the corresponding picture area of the no-parking area B. At this time, the scene corresponding to the no-parking area B is the target scene.

[0067] In the embodiments of the present application, the image information and depth information of the target scene refer to the image information of the objects in the target scene and the depth information corresponding to the objects. That is, there is a corresponding relationship between the obtained image information and depth information.

[0068] Optionally, considering that compared with grayscale images, color images (such as RGB-format images) contain more information, so when obtaining the image information of the target scene, the RGB-format image information of the target scene can be obtained to improve the accuracy of the obtained image information of the target scene.

[0069] Optionally, the above-mentioned image information of the target scene and the depth information corresponding to the above-mentioned image information can be obtained through a dual-mode acquisition device.

[0070] Specifically, when the image information adopted is RGB-format image information, the above-mentioned dual-mode acquisition device can be an RGB-D dual-mode acquisition device, which is used to acquire depth information and RGB-format image information corresponding to the depth information. When using this RGB-D dual-mode acquisition device for data acquisition, paired data can be acquired, and each pair of data includes an RGB-format image information and a corresponding depth information.

[0071] Optionally, the image information and depth information of the target scene can also be obtained in the following manner: obtain the above-mentioned image information of the target scene through an image acquisition device, and, output the depth information corresponding to the above-mentioned image information of the target scene through a pre-trained depth estimation model, where the above-mentioned pre-trained depth estimation model is trained according to the image information for training and the depth information for training of the above-mentioned target scene.

[0072] Specifically, a limited number of image information and corresponding depth information in the target scene are obtained. Of course, the image information and corresponding depth information can also be obtained by a dual-mode acquisition device. The difference is that the image information and depth information obtained here are used to train an initial depth estimation model, so the image information and depth information obtained here are less. After obtaining the image information and depth information of the target scene in a limited number, an open-source bullet screen depth estimation model can be selected as the initial depth estimation model, and the obtained limited number of image information and depth information of the target scene are used to train the initial depth estimation model to obtain a pre-trained depth estimation model. Continuously obtain the image information in the target scene, and input the image information into the pre-trained depth estimation model to obtain the depth information corresponding to the image information output by the pre-trained depth estimation model.

[0073] Since the cost of simultaneously acquiring image information and depth information is relatively high, therefore, some image information and depth information are first acquired to train a pre-trained depth estimation model. Subsequently, there is no need to simultaneously acquire image information and depth information. Only after acquiring new image information, the new depth information can be predicted according to the pre-trained depth estimation model. Through the above processing, it is beneficial to reduce the cost consumed by acquiring information.

[0074] S12. Generate a three-dimensional model of the target scene according to the above image information and the above depth information.

[0075] Specifically, by obtaining the internal parameters (such as focal length and optical center coordinates) and external parameters (such as the position and orientation of the imaging device represented by a selection matrix and a translation vector) of the imaging device that obtains the image information, the two-dimensional coordinates of the image information are mapped to two components of three-dimensional coordinates, and combined with the depth information corresponding to the image information (i.e., the third component of the three-dimensional coordinates), the corresponding three-dimensional coordinates are obtained. Surface reconstruction is performed according to each three-dimensional coordinate to obtain a three-dimensional model of the target scene.

[0076] Optionally, the above three-dimensional model can be represented by a point cloud model or an implicit representation. For example, an implicit representation of a three-dimensional scene using the 3D Gaussian Splatting technique is adopted. Among them, the implicit representation of a three-dimensional scene using the 3D Gaussian Splatting technique represents the point cloud in the target scene using a Gaussian distribution. Each Gaussian point not only contains position information but also a covariance matrix describing the shape and size of the point cloud distribution, and also contains attributes such as color and transparency. By adopting the implicit representation of a three-dimensional scene using the 3D Gaussian Splatting technique, it is beneficial to improve the rendering speed and the flexibility of representing the target scene.

[0077] Optionally, to improve the integrity and accuracy of the generated three-dimensional model, image information and depth information at different positions and / or from different perspectives can be obtained, and then the corresponding three-dimensional model can be generated based on this image information and depth information. That is, when the image information and depth information are obtained by a photographing device at different positions and / or from different perspectives of the target scene, generating the three-dimensional model of the target scene based on the above image information and the above depth information includes:

[0078] A1. Determine the three-dimensional pose of the photographing device in the target coordinate system according to each of the above image information and the above depth information to obtain the corresponding three-dimensional pose of the photographing device, where the target coordinate system is the coordinate system where the target scene is located.

[0079] A2. Generate the three-dimensional model of the target scene according to each of the above three-dimensional poses of the photographing device, the above image information, and the above depth information.

[0080] In the embodiment of the present application, when the photographing device obtains the image information and depth information, the Structure from Motion (SfM) algorithm is used to analyze the motion changes of objects in each image information, and infer the three-dimensional structure of the target scene and the motion trajectory of the photographing device during the photographing process. This SfM algorithm uses local feature matching between images to solve the absolute pose of the photographing device in the target coordinate system (i.e., the global unified coordinate system), thereby realizing three-dimensional reconstruction. Specifically, feature points (such as corner points, edge points, etc.) can be extracted from each image information, and the feature points extracted from different image information are matched to determine the corresponding relationship of the same scene point from different perspectives. According to the corresponding relationship of the matched feature points, the motion parameters of the photographing device are estimated, that is, the absolute pose of the photographing device in the target coordinate system is estimated. After estimating the absolute pose of the photographing device in the target coordinate system, the three-dimensional coordinates of the matched feature points in the target coordinate system can be calculated according to the pose of the photographing device, the image information, and the depth information, and the three-dimensional coordinates of each matched feature point in the target coordinate system constitute the three-dimensional model of the target scene.

[0081] Since the amount of information of the target scene contained in the image information and depth information at different positions and / or from different perspectives is greater than the amount of information of the image information and depth information at the same position, the three-dimensional model of the target scene generated according to the image information and depth information at different positions and / or from different perspectives is more complete and accurate.

[0082] S13. Obtain the information of the target to be measured.

[0083] Among them, the information of the target to be measured refers to the information of the target to be measured, which may include the name of the target to be measured, may also include color information, may also include the coordinate position information of the target to be measured, and of course, may also include other information, which is not limited here.

[0084] In the embodiments of the present application, the target to be measured is the target that needs to be detected, which may be a static object or a dynamic object, and is not limited here.

[0085] S14. Based on the three-dimensional model of the above target scene and the above information of the target to be measured, generate a simulation video of the target to be measured in the above target scene.

[0086] Specifically, neural rendering technology can be used to generate the simulation video. The scene in the simulation video belongs to the target scene, and the target scene includes the target to be measured. For example, when the target to be measured is a vehicle, the video frames in the simulation video include the image information corresponding to the vehicle. If the information of the target to be measured (i.e., the information of the vehicle) does not include the coordinate position information of the vehicle, the position of the vehicle in the simulation video may be random; if the information of the vehicle includes the coordinate position information of the vehicle, the position of the vehicle in the simulation video is the position corresponding to the coordinate position information.

[0087] Optionally, each video frame in the generated simulation video includes the target to be measured, and the performance information of the target to be measured is different in at least two video frames. Among them, the performance information here includes coordinate position information, and may also include shape information and / or size information. Since the setting of the trigger line is related to the performance information of the target to be measured in the video frame, when each video frame in the generated simulation video includes the target to be measured and the performance information of the target to be measured is different in at least two video frames, it is beneficial to subsequently check the effects generated by attempting to set the trigger line at different positions in the video frame, and thus beneficial to improving the accuracy of the finally set trigger line.

[0088] S15. Iteratively optimize the initially set trigger line according to the above simulation video to obtain the optimized trigger line.

[0089] Among them, the initially set trigger line can be set according to experience or randomly. Optionally, the trigger line here can be a straight line (or line segment or curve), or multiple straight lines (or line segments or curves), and its form is related to the specific usage scenario and requirements, which is not limited here.

[0090] After the initial setting is performed on the trigger line, the initial position information and size information corresponding to the trigger line will be obtained. Subsequently, after detecting that the target to be measured appears at the position where the trigger line is located, the corresponding action will be triggered. If it is determined that the effect after the trigger action does not meet the preset requirements, the initially set trigger line will be adjusted, and then the target to be measured will be detected according to the adjusted trigger line to determine whether the effect meets the preset requirements when the action is triggered according to the adjusted trigger line. Of course, if it is determined that the effect after the trigger action meets the preset requirements, there is no need to adjust the current trigger line, that is, the iteration is stopped, and the current trigger line is the optimized trigger line. Specifically, during the playback of the simulation video, it is detected whether the target to be measured appears at the position of the initially set trigger line. If so, the action corresponding to the trigger line is executed. For example, when the action corresponding to the trigger line is a capture action, when the target to be measured appears at the position of the initially set trigger line, the video frame currently being played is captured; when the action corresponding to the trigger line is to control the lifting or lowering of the barrier gate, when the target to be measured appears at the position of the initially set trigger line, the barrier gate is controlled to be lifted or lowered; when the action corresponding to the trigger line is to output a warning sound, when the target to be measured appears at the position of the initially set trigger line, a warning sound is output.

[0091] S16. Set the trigger line of the target scene according to the optimized trigger line above.

[0092] Specifically, use the optimized trigger line as the trigger line of the target scene. For example, set the position and size of the trigger line for the target scene according to the position and size of the optimized trigger line in the video frame.

[0093] In the embodiment of the present application, according to the image information and depth information of the target scene, a three-dimensional model of the target scene is generated, and according to the three-dimensional model and the obtained information of the target to be measured, a simulation video of the target to be measured in the target scene is generated. Since the simulation video is generated according to the image information and depth information of the target scene, the simulation video can reflect the relevant information of the target scene, so that the user can view the relevant information of the target scene from the simulation video without reaching the specific scene. In addition, since the simulation video also includes the target to be measured, when the initially set trigger line is iteratively optimized according to the generated simulation video, it is equivalent to detecting the target to be measured with different trigger lines, which is beneficial to determining a better trigger line as the trigger line of the target scene. That is, when setting the trigger line by the above method, since the user does not need to reach the site for setting, the setting efficiency of the trigger line is improved. In addition, since the trigger line of the target scene is set according to the iteratively optimized trigger line, the influence of personal experience on the set trigger line is reduced, which is beneficial to improving the accuracy of the set trigger line.

[0094] In some embodiments, S14, based on the three-dimensional model of the above target scenario and the above information of the target to be measured, generate a simulation video of the target to be measured in the above target scenario, including:

[0095] B1. Determine the image features at a specified angle in the three-dimensional model of the above target scenario to obtain background image features.

[0096] Among them, the specified angle generally includes azimuth, pitch, and roll angles.

[0097] Specifically, when the three-dimensional model of the target scenario is represented by multiple three-dimensional Gaussian spheres, rasterization technology can be used to project the Gaussian spheres at the specified angle onto the image plane to obtain background image features. Further, after projecting the Gaussian spheres at the specified angle onto the image plane, a diffusion model can be used to render the projection result, and the rendered features are the above-mentioned background image features. Since the diffusion model can simulate complex lighting and shadow effects, and can generate realistic materials and textures, after rendering the projection result through the diffusion model, the generated image features can be more physically reasonable and visually realistic.

[0098] Optionally, the above image features can be the image features within the specified angle and a preset field of view. For example, a field of view range and a specified angle are preset in advance, and the image features within the specified angle and the preset field of view are determined in the three-dimensional model of the target scenario to obtain background image features, so as to improve the accuracy of the obtained background image features.

[0099] B2. Generate a simulation video of the target to be measured in the above target scenario based on the above background image features, the three-dimensional model of the above target scenario, and the above information of the target to be measured.

[0100] Specifically, use the background image features as the image features of the background of the target to be measured in the video frame to dynamically and controllably generate a simulation video from the current perspective of the three-dimensional model of the target scenario.

[0101] Optionally, use the background features, three-dimensional model, and target information in the training set to train an initial text-to-image generation model to obtain a trained text-to-image generation model. When the trained text-to-image generation model inputs the background image features, the three-dimensional model of the target scenario (such as the implicit expression of the three dimensions of the target scenario), and the information of the target to be measured, the text-to-image generation model will output the corresponding simulation video. Since the trained text-to-image generation model can output the corresponding simulation video without too much manual intervention, when generating the simulation video through the above method, it is beneficial to improve the output efficiency of the simulation video.

[0102] In some embodiments, the above-mentioned information of the target to be measured includes the text description information and coordinate position information of the above-mentioned target to be measured. Based on the above-mentioned background image features, the three-dimensional model of the target scene, and the above-mentioned information of the target to be measured, generating a simulation video of the target to be measured in the above-mentioned target scene includes:

[0103] B21. Extract the features of the text description information of the above-mentioned target to be measured to obtain text features, and extract the features of the above-mentioned coordinate position information to obtain digital features.

[0104] Among them, the text description information of the target to be measured refers to the information that describes the target to be measured in words. When the text description information is "a black car", the target to be measured is "black car". And when the text description information is "a person", the target to be measured is "person".

[0105] In the embodiments of the present application, for the text description information of the target to be measured, the text description information can be converted into a feature vector that can represent its semantic information through a text encoder, and this feature vector is used as the text feature of the text description information, which can be a feature vector with a fixed length. For the coordinate position information, since the coordinate position information is discrete digital information, therefore, the discrete digital information can be mapped into Fourier features, and the obtained Fourier features are used as the digital features corresponding to the coordinate position information.

[0106] Optionally, considering that the simulation video usually includes multiple video frames, therefore, in order to make the targets to be measured in each video frame have differences in position, the number of the above-mentioned coordinate position information can be multiple, and its number can be equal to the number of video frames in the simulation video.

[0107] B22. Align the above-mentioned text features, the above-mentioned digital features, and the above-mentioned background image features to obtain the aligned above-mentioned text features, the aligned above-mentioned digital features, and the aligned above-mentioned background image features.

[0108] Specifically, use a multi-layer perceptron to map the text features and digital features to the same dimension as the background image features to achieve the alignment of the text features, digital features, and background image features.

[0109] Optionally, when the background image features are continuous high-dimensional information, the extracted text features and digital features are also encoded high-dimensional features to facilitate the subsequent implementation of feature alignment.

[0110] B23. Generate the above-mentioned simulation video of the target to be measured in the above-mentioned target scene according to the aligned above-mentioned text features, the aligned above-mentioned digital features, the aligned above-mentioned background image features, and the three-dimensional model of the above-mentioned target scene.

[0111] Specifically, when generating a simulation video, the aligned text features, aligned numerical features, and aligned background image features can be interactively learned by using the Cross-Attention Mechanism to improve the effectiveness of information fusion. Among them, the Cross-Attention Mechanism is an attention calculation method that enables information interaction between two different sequences. It allows one sequence to "attend to" relevant parts of the other sequence during generation or processing, thereby obtaining richer context information. The core of this mechanism lies in calculating the similarity between the two sequences and assigning weights to each element based on the similarity to achieve effective information fusion.

[0112] In the embodiments of the present application, since the text features, numerical features, and background image features belong to features of different modalities, aligning these features of different modalities is equivalent to unifying these features of different modalities into the same semantic space, that is, enabling cross-modal information to be more effectively fused and understood, thereby making the generated simulation video more accurate.

[0113] In some embodiments, the above simulation video is used to reflect the movement process of the above-mentioned target to be measured within a continuous time. Here, the continuous time is greater than 0 and not greater than the duration of the simulation video. When the purpose of setting the trigger line is to capture images, the above S15 iteratively optimizes the initially set trigger line according to the above simulation video to obtain an optimized trigger line, including:

[0114] C1. During the playback of the above simulation video, detect whether the above-mentioned target to be measured in the currently played video frame is on the initially set above trigger line.

[0115] Since the simulation video is used to reflect the movement process of the target to be measured within a continuous time, there are at least two video frames in the simulation video with different positions of the target to be measured included. Therefore, it is necessary to detect whether the currently played video frame contains the target to be measured, and when it is detected that the currently played video frame contains the target to be measured, whether the position of the target to be measured in the image frame belongs to the position of the initially set trigger line. If so, it is determined that the target to be measured in the currently played video frame is on the initially set trigger line.

[0116] Optionally, the following method can be used to detect whether the target to be measured is on the trigger line:

[0117] (1) If the trigger line can be expressed as a mathematical equation, the perpendicular distance from the center point of the target to be measured to the trigger line can be calculated. If the perpendicular distance is less than a preset distance threshold, it can be determined that the target to be measured is on the trigger line. If the center point of the target to be measured satisfies the mathematical equation corresponding to the trigger line, it can also be determined that the target to be measured is on the trigger line.

[0118] (2) If the trigger line and the target to be measured can be represented as bounding boxes, the intersection over union (IoU) of the bounding box corresponding to the trigger line and the bounding box of the target to be measured can be calculated. If the IoU is greater than a preset IoU threshold, it can be determined that the target to be measured is on the trigger line.

[0119] Of course, other methods can also be used for judgment, which will not be elaborated here.

[0120] C2. If the above-mentioned target to be measured in the currently played video frame is on the initially set above-mentioned trigger line, the above-mentioned simulation video is captured to obtain a captured image.

[0121] Specifically, if it is detected that the target to be measured is on the initially set trigger line in the currently played video frame, the currently played simulation video is captured to obtain a captured image. Optionally, the captured image can be the currently played video frame, or it may be the next video frame in the currently played video frame.

[0122] C3. Analyze the above-mentioned target to be measured in the above-mentioned captured image to obtain an analysis result.

[0123] Optionally, the analysis here can include the analysis of the clarity of the target to be measured.

[0124] Specifically, considering that the purpose of setting the trigger line is to capture the target to be measured on the trigger line, therefore, in the captured image, the clearer the target to be measured, the higher the probability of analyzing the target to be measured from the captured image. At this time, the probability that the set trigger line meets the user's requirements is higher. Therefore, the analysis result here can include the clarity of the target to be measured.

[0125] Optionally, the analysis here can also include the analysis of the size of the target to be measured.

[0126] Specifically, considering that in the captured image, the more area occupied by the target to be measured, the more information about the target to be measured the captured image contains, and the higher the probability of analyzing the target to be measured from the captured image. At this time, the probability that the set trigger line meets the user's requirements is higher. Therefore, the analysis result here can include the size of the target to be measured. It should be noted that if the analysis of the target to be measured is an analysis of the target to be measured itself, the size of the target to be measured refers to the size of the target to be measured, and if the analysis of the target to be measured is an analysis of a partial area of the target to be measured, the size of the target to be measured refers to the size of the partial area of the target to be measured.

[0127] C4. If the above-mentioned analysis result does not meet the preset requirements, the initially set above-mentioned trigger line is adjusted to obtain the adjusted above-mentioned trigger line.

[0128] Among them, the preset requirements here are set according to actual needs. Optionally, the above preset requirements may include that the clarity is greater than a preset clarity threshold, and may also include that the ratio of the target to be measured to the captured image is greater than a preset ratio threshold, and so on.

[0129] In the embodiments of the present application, the trigger line of the initial setting is adjusted, including adjusting at least one of the position, size, and angle of the trigger line of the initial setting. For example, when the analysis result indicates that the clarity does not meet the requirements, the position parameter of the trigger line can be adjusted so that its position in the image is closer to the focal plane of the shooting device. Since when the position of the trigger line is close to the focal plane of the shooting device, the target to be measured will be on the trigger line when it is closer to the focal plane of the shooting device, the shooting device can capture a clearer target to be measured according to the adjusted trigger line.

[0130] C5. Take the adjusted trigger line as the trigger line of the new initial setting, and return to the step of detecting whether the target to be measured in the currently played video frame is on the trigger line of the initial setting and subsequent steps during the playback of the simulation video until the analysis result meets the above preset requirements.

[0131] After stopping the iteration, the obtained trigger line is the optimized trigger line.

[0132] Specifically, take the adjusted trigger line as the trigger line of the new initial setting, continue to play the simulation video, and detect whether the target to be measured is on the trigger line of the new initial setting to determine whether to stop optimizing the trigger line.

[0133] In the embodiments of the present application, since the analysis result obtained by analyzing the target to be measured in the captured image is used to select whether to iterate the set trigger line, such as selecting to iterate the set trigger line when the analysis result does not meet the preset requirements until the analysis result meets the preset requirements. Therefore, for the finally set trigger line according to the above method, when the target to be measured is on this trigger line, the analysis result of the captured image will meet the preset requirements.

[0134] In some embodiments, when the purpose of setting the trigger line is to control the lifting or lowering of the gate rod, the above S15, iteratively optimize the trigger line of the initial setting according to the above simulation video to obtain the optimized trigger line, including:

[0135] D1. During the playback of the simulation video, detect whether the target to be measured in the currently played video frame is on the trigger line of the initial setting.

[0136] Among them, for the process of detecting whether the target to be measured is on the trigger line of the initial setting, please refer to C1 above and will not be elaborated here.

[0137] D2. If the above-mentioned target to be measured in the currently played video frame is on the above-mentioned trigger line set initially, control the barrier gate rod to lift or lower.

[0138] Specifically, if the barrier gate rod is in the lowered state when no target to be measured is detected, then when it is detected that the target to be measured is on the trigger line, control the barrier gate rod to lift; if the barrier gate rod is in the lifted state when no target to be measured is detected, then when it is detected that the target to be measured is on the trigger line, control the barrier gate rod to lower.

[0139] D3. Statistically calculate the duration from the moment when it is detected that the above-mentioned target to be measured is on the above-mentioned trigger line set initially to the moment when the control of the above-mentioned barrier gate rod is completed, and obtain T1.

[0140] Since the barrier gate rod will be controlled to lift or lower after it is detected that the target to be measured is on the trigger line set initially, therefore, the above-mentioned T1 is related to the detection frame rate of the video frame and the response speed of the barrier gate rod. When the detection frame rate of the video frame is smaller and the response speed of the barrier gate rod is slower, the above-mentioned T1 is larger; on the contrary, when the detection frame rate of the video frame is larger and the response speed of the barrier gate rod is faster, the above-mentioned T1 is smaller.

[0141] D4. If T1 is greater than the preset duration threshold T2, then adjust the position parameter and / or the angle of the above-mentioned trigger line set initially to make its position in the image away from the focal plane of the shooting device, and obtain the adjusted above-mentioned trigger line, where T2 is the duration required for the target to be measured to move from the position of the trigger line to the barrier gate rod at the preset moving speed.

[0142] Among them, the above-mentioned preset duration threshold T2 is set according to the actual situation. For example, when the target to be measured is a vehicle, the above-mentioned preset moving speed is relatively large, and when the target to be measured is a pedestrian, the above-mentioned preset moving speed is relatively small.

[0143] In the embodiment of the present application, when adjusting the position parameter and / or the angle of the trigger line to make its position in the image away from the focal plane of the shooting device, the shooting device will be able to detect the target to be measured earlier, so as to be able to control the barrier gate rod earlier, and to minimize the impact of the barrier gate rod on the passage of the target to be measured.

[0144] Optionally, if T1 is less than T2, then the position parameter and / or the angle of the trigger line set initially can be adjusted to make its position in the image close to the focal plane of the shooting device, and obtain the adjusted trigger line. Of course, it is also possible to adjust the position parameter of the trigger line set initially to make its position in the image close to the focal plane of the shooting device and obtain the adjusted trigger line after it is determined that T1 is less than T2 and the difference between T2 and T1 is greater than the preset difference threshold.

[0145] D5. Use the adjusted trigger line as the newly initially set trigger line, and return to the step of detecting whether the target to be measured in the currently played video frame is on the initially set trigger line during the playback of the simulation video and subsequent steps until the obtained T is not greater than the preset duration threshold T2.

[0146] After stopping the iteration, the obtained trigger line is the optimized trigger line.

[0147] In the embodiment of the present application, when controlling the gate rod through the trigger line, after determining that T1 is greater than the preset duration threshold T2, the position parameter of the initially set trigger line and / or the angle of the trigger line are adjusted so that its position in the image is far from the focal plane of the shooting device, and the adjusted trigger line is obtained. Since T1 is the duration from the moment when the target to be measured is detected on the initially set trigger line to the moment when the control of the gate rod is completed, and T2 is the duration required for the target to be measured to move from the position of the trigger line to the gate rod at the preset moving speed, that is, when T1 is greater than T2, it indicates that after the target to be measured moves to the gate rod, the control of the gate rod has not ended yet. At this time, making the position of the trigger line in the image far from the focal plane of the shooting device is equivalent to starting to control the gate rod at a place farther from the shooting device. Therefore, it is beneficial to reduce the probability that the control of the gate rod has not ended after the target to be measured reaches the gate rod, thereby improving the good experience of the target to be measured.

[0148] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0149] Corresponding to the method for setting the trigger line described in the above embodiments, Figure 2 The structural block diagram of the trigger line setting device provided by the embodiment of the present application is shown. For the sake of convenience of description, only the parts related to the embodiment of the present application are shown.

[0150] Refer to Figure 2 , the trigger line setting device can be applied to a network device, including: a target scene information acquisition module 21, a three-dimensional model generation module 22 of the target scene, a target to be measured information acquisition module 23, a simulation video generation module 24, a trigger line optimization module 25, and an optimized trigger line setting module 26. Among them:

[0151] The target scene information acquisition module 21 is used to acquire the image information and depth information of the target scene;

[0152] The three-dimensional model generation module 22 of the target scene is used to generate the three-dimensional model of the target scene according to the image information and the depth information;

[0153] The target information acquisition module 23 to be measured is used to acquire the target information to be measured;

[0154] The simulation video generation module 24 is used to generate a simulation video of the target to be measured in the target scenario based on the three-dimensional model of the target scenario and the target information to be measured;

[0155] The trigger line optimization module 25 is used to iteratively optimize the initially set trigger line according to the simulation video to obtain an optimized trigger line;

[0156] The optimized trigger line setting module 26 is used to set the trigger line of the target scenario according to the optimized trigger line.

[0157] In the embodiment of the present application, a three-dimensional model of the target scenario is generated according to the image information and depth information of the target scenario, and a simulation video of the target to be measured in the target scenario is generated according to the three-dimensional model and the acquired target information to be measured. Since the simulation video is generated according to the image information and depth information of the target scenario, the simulation video can reflect the relevant information of the target scenario, so that the user can view the relevant information of the target scenario from the simulation video without reaching the specific scenario. In addition, since the simulation video also includes the target to be measured, when iteratively optimizing the initially set trigger line according to the generated simulation video, it is equivalent to detecting the target to be measured with different trigger lines, which is conducive to determining a better trigger line as the trigger line of the target scenario. That is, when setting the trigger line by the above method, since the setting can be performed without the user arriving at the scene, the setting efficiency of the trigger line is improved. In addition, since the trigger line of the target scenario is set according to the iteratively optimized trigger line, the influence of personal experience on the set trigger line is reduced, which is conducive to improving the accuracy of the set trigger line.

[0158] In some embodiments, the target scenario information acquisition module 21 includes:

[0159] The first depth information acquisition unit is used to acquire the image information of the target scenario and the depth information corresponding to the image information through a dual-mode acquisition device.

[0160] Or,

[0161] The second depth information acquisition unit is used to acquire the image information of the target scenario through an image acquisition device, and output the depth information corresponding to the image information of the target scenario through a pre-trained depth estimation model, where the pre-trained depth estimation model is trained according to the image information for training and the depth information for training of the target scenario.

[0162] In some embodiments, the above-mentioned target scenario information acquisition module 21 is specifically configured to:

[0163] Obtain the image information and depth information of the above-mentioned target scenario through a photographing device at different positions and / or different perspectives of the above-mentioned target scenario.

[0164] Correspondingly, the above-mentioned three-dimensional model generation module 22 of the target scenario includes:

[0165] A three-dimensional pose determination unit of the photographing device, configured to determine the three-dimensional pose of the photographing device in a target coordinate system according to each of the above-mentioned image information and the above-mentioned depth information, to obtain the corresponding three-dimensional pose of the photographing device, where the above-mentioned target coordinate system is the coordinate system where the above-mentioned target scenario is located.

[0166] A three-dimensional model construction unit, configured to generate a three-dimensional model of the above-mentioned target scenario according to each of the above-mentioned three-dimensional poses of the photographing device, the above-mentioned image information, and the above-mentioned depth information.

[0167] In some embodiments, the above-mentioned simulation video generation module 24 includes:

[0168] A background image feature generation unit, configured to determine the image features at a specified angle in the three-dimensional model of the above-mentioned target scenario to obtain background image features.

[0169] A simulation model generation unit, configured to generate a simulation video of the target to be measured in the above-mentioned target scenario based on the above-mentioned background image features, the three-dimensional model of the above-mentioned target scenario, and the information of the target to be measured.

[0170] In some embodiments, the information of the target to be measured includes the text description information and coordinate position information of the target to be measured, and the simulation model generation unit is specifically configured to:

[0171] Extract the features of the text description information of the above-mentioned target to be measured to obtain text features, and extract the features of the above-mentioned coordinate position information to obtain digital features;

[0172] Perform alignment processing on the above-mentioned text features, the above-mentioned digital features, and the above-mentioned background image features to obtain the aligned above-mentioned text features, the aligned above-mentioned digital features, and the aligned above-mentioned background image features;

[0173] Generate a simulation video of the above-mentioned target to be measured in the above-mentioned target scenario according to the aligned above-mentioned text features, the aligned above-mentioned digital features, the aligned above-mentioned background image features, and the three-dimensional model of the above-mentioned target scenario.

[0174] In some embodiments, the above-mentioned simulation video is used to reflect the process of the target to be measured moving within a continuous time, and the trigger line optimization module 25 is specifically configured to:

[0175] During the playback of the above simulation video, detect whether the above target to be measured in the currently played video frame is on the above trigger line set initially;

[0176] If the above target to be measured in the currently played video frame is on the above trigger line set initially, perform a capture process on the above simulation video to obtain a captured image;

[0177] Analyze the above target to be measured in the above captured image to obtain an analysis result;

[0178] If the above analysis result does not meet the preset requirements, adjust the above trigger line set initially to obtain the adjusted above trigger line;

[0179] Use the adjusted trigger line as the newly set initial above trigger line, and return to the step of detecting whether the above target to be measured in the currently played video frame is on the above trigger line set initially during the playback of the above simulation video and subsequent steps until the above analysis result meets the above preset requirements.

[0180] In some embodiments, the above simulation video is used to reflect the movement process of the above target to be measured within a continuous time, and the trigger line optimization module 25 is specifically configured to:

[0181] During the playback of the above simulation video, detect whether the above target to be measured in the currently played video frame is on the above trigger line set initially;

[0182] If the above target to be measured in the currently played video frame is on the above trigger line set initially, control the barrier gate pole to lift or lower;

[0183] Statistically calculate the duration from the moment when it is detected that the above target to be measured is on the above trigger line set initially to the moment when the control of the above barrier gate pole is completed to obtain T1;

[0184] If T1 is greater than the preset duration threshold T2, adjust the position parameter and / or the angle of the above trigger line set initially so that its position in the image is away from the focal plane of the shooting device to obtain the adjusted above trigger line, where T2 is the duration required for the target to be measured to move from the position of the above trigger line to the barrier gate pole at a preset moving speed;

[0185] Use the adjusted trigger line as the newly set initial above trigger line, and return to the step of detecting whether the above target to be measured in the currently played video frame is on the above trigger line set initially during the playback of the above simulation video and subsequent steps until the obtained T is not greater than the above preset duration threshold T2.

[0186] It should be noted that for the content such as information interaction and execution process between the above-mentioned devices / units, since it is based on the same concept as the method embodiment of the present application, for its specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details are not described herein again.

[0187] Figure 3 It is a schematic structural diagram of a network device provided by an embodiment of the present application. As Figure 3 shown, the network device 3 of this embodiment includes: at least one processor 30 ( Figure 3 only one processor is shown in the figure), a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30. When the processor 30 executes the computer program 32, it implements the steps in any of the above-mentioned method embodiments.

[0188] The network device 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art can understand that Figure 3 this is only an example of the network device 3 and does not constitute a limitation on the network device 3. It may include more or fewer components than shown in the figure, or combine some components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0189] The so-called processor 30 may be a central processing unit (CPU), and this processor 30 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or this processor may also be any conventional processor, etc.

[0190] The memory 31 may be an internal storage unit of the network device 3 in some embodiments, such as a hard disk or memory of the network device 3. The memory 31 may also be an external storage device of the network device 3 in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the network device 3. Further, the memory 31 may also include both the internal storage unit and the external storage device of the network device 3. The memory 31 is used to store an operating system, application programs, a BootLoader, data, and other programs, such as the program code of the computer program. The memory 31 may also be used to temporarily store data that has been output or will be output.

[0191] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above-mentioned functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0192] An embodiment of this application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. When the processor executes the computer program, the steps in any of the foregoing method embodiments are implemented.

[0193] An embodiment of this application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in any of the foregoing method embodiments can be implemented.

[0194] An embodiment of this application provides a computer program product. When the computer program product runs on a network device, the network device can be made to execute the steps in any of the foregoing method embodiments.

[0195] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present application, a computer program can be used to instruct relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / network device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0196] In the above embodiments, the descriptions of the various embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0197] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0198] In the embodiments provided in this application, it should be understood that the disclosed device / network device and method can be implemented in other ways. For example, the device / network device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0199] The unit described as a separation component may or may not be physically separated. The component displayed as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0200] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for setting a trigger line, characterized in that, Including: Obtaining the image information and depth information of the target scene; Generating a three-dimensional model of the target scene according to the image information and the depth information; Obtaining the information of the target to be measured; Generating a simulation video of the target to be measured in the target scene based on the three-dimensional model of the target scene and the information of the target to be measured; Iteratively optimizing the initially set trigger line according to the simulation video to obtain an optimized trigger line; Setting the trigger line of the target scene according to the optimized trigger line.

2. The method for setting a trigger line according to claim 1, wherein, The obtaining the image information and depth information of the target scene includes: Obtaining the image information of the target scene and the depth information corresponding to the image information through a dual-mode acquisition device; Or, Obtaining the image information of the target scene through an image acquisition device, and outputting the depth information corresponding to the image information of the target scene through a pre-trained depth estimation model, where the pre-trained depth estimation model is trained according to the image information for training and the depth information for training of the target scene.

3. The method for setting a trigger line according to claim 1, wherein The obtaining the image information and depth information of the target scene includes: Obtaining the image information and depth information of the target scene through a photographing device at different positions and / or different perspectives of the target scene; The generating a three-dimensional model of the target scene according to the image information and the depth information includes: Determining the three-dimensional pose of the photographing device in the target coordinate system according to each piece of the image information and the depth information to obtain the corresponding three-dimensional pose of the photographing device, where the target coordinate system is the coordinate system where the target scene is located; Generating a three-dimensional model of the target scene according to each piece of the three-dimensional pose of the photographing device, the image information, and the depth information.

4. The method for setting the trigger line according to claim 1, wherein The generating a simulation video of the target to be measured in the target scene based on the three-dimensional model of the target scene and the information of the target to be measured includes: Determining the image features at a specified angle in the three-dimensional model of the target scene to obtain background image features; Generating a simulation video of the target to be measured in the target scene based on the background image features, the three-dimensional model of the target scene, and the information of the target to be measured.

5. The method for setting the trigger line according to claim 4, characterized in that, The information of the target to be measured includes the text description information and coordinate position information of the target to be measured. The generating a simulation video of the target to be measured in the target scene based on the background image features, the three-dimensional model of the target scene, and the information of the target to be measured includes: Extracting the features of the text description information of the target to be measured to obtain text features, and extracting the features of the coordinate position information to obtain digital features; Performing alignment processing on the text features, the digital features, and the background image features to obtain the aligned text features, the aligned digital features, and the aligned background image features; Generating a simulation video of the target to be measured in the target scene according to the aligned text features, the aligned digital features, the aligned background image features, and the three-dimensional model of the target scene.

6. The method for setting the trigger line according to any one of claims 1 to 5, characterized in that The simulation video is used to reflect the movement process of the target to be measured within continuous time. Iteratively optimizing the initially set trigger line according to the simulation video to obtain an optimized trigger line includes: During the playback of the simulation video, detecting whether the target to be measured in the currently played video frame is on the initially set trigger line; If the target to be measured in the currently played video frame is on the initially set trigger line, performing a capture process on the simulation video to obtain a captured image; Analyzing the target to be measured in the captured image to obtain an analysis result; If the analysis result does not meet the preset requirements, adjusting the initially set trigger line to obtain an adjusted trigger line; Taking the adjusted trigger line as the newly initially set trigger line, and returning to the step of detecting whether the target to be measured in the currently played video frame is on the initially set trigger line during the playback of the simulation video and subsequent steps until the analysis result meets the preset requirements.

7. The method for setting the trigger line according to any one of claims 1 to 5, characterized in that The simulation video is used to reflect the movement process of the target to be measured within continuous time. Iteratively optimizing the initially set trigger line according to the simulation video to obtain an optimized trigger line includes: During the playback of the simulation video, detecting whether the target to be measured in the currently played video frame is on the initially set trigger line; If the target to be measured in the currently played video frame is on the initially set trigger line, controlling the barrier gate rod to lift or lower; Statistical the duration from the moment when it is detected that the target to be measured is on the initially set trigger line to the moment when the control of the barrier gate rod is completed to obtain T1; If T1 is greater than a preset duration threshold T2, adjusting the position parameter and / or the angle of the initially set trigger line so that its position in the image is away from the focal plane of the imaging device to obtain an adjusted trigger line, where T2 is the duration required for the target to be measured to move from the position of the trigger line to the barrier gate rod at a preset moving speed; Taking the adjusted trigger line as the newly initially set trigger line, and returning to the step of detecting whether the target to be measured in the currently played video frame is on the initially set trigger line during the playback of the simulation video and subsequent steps until the obtained T is not greater than the preset duration threshold T2.

8. A triggering wire setting device, characterized in that, Including: A target scene information acquisition module, configured to acquire image information and depth information of a target scene; A three-dimensional model generation module of the target scene, configured to generate a three-dimensional model of the target scene according to the image information and the depth information; A target to be measured information acquisition module, configured to acquire information of the target to be measured; A simulation video generation module, configured to generate a simulation video of the target to be measured in the target scene based on the three-dimensional model of the target scene and the information of the target to be measured; A trigger line optimization module, configured to iteratively optimize the initially set trigger line according to the simulation video to obtain an optimized trigger line; The optimized trigger line setting module is used to set the trigger line of the target scenario according to the optimized trigger line.

9. A network device, 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, the method described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the method described in any one of claims 1 to 7 is implemented.

11. A computer program product, characterized in that, It includes a computer program, and when the computer program is run, the method described in any one of claims 1 to 7 is executed.

Images