Entrance guard advertisement pushing method based on 3D imaging

Abstract

According to the access control advertisement pushing method based on 3D imaging, information such as the position, the speed and the orientation angle of a user can be accurately obtained through 3D imaging data collection and user detection and tracking. The advertisement 3D model is subjected to translation, rotation and zooming adjustment according to the information, so that the advertisement is always located in the optimal view area in front of the user, personalized advertisement pushing according to the specific condition of the user is realized, and the pertinence and attraction of the advertisement are improved. A user can send out an instruction through a gesture or AI glasses, and switching or link skipping of advertisement content is achieved. According to the interaction mode, the user can actively participate in the advertisement experience, the participation degree of the user and the attention degree of the advertisement are increased, and the propagation effect of the advertisement can be improved. And by adopting a 3D holographic display technology, the adaptive advertisement 3D model is projected to a holographic display and displayed as a free space 3D advertisement, so that the attention of a user is more easily attracted, and the memory degree of the advertisement is improved.

Description

Technical Field

[0001] This invention relates to the field of access control management and application technology, and in particular to an access control advertising push method based on 3D imaging, as well as an access control advertising push device, electronic device, and computer-readable storage medium based on 3D imaging. Background Technology

[0002] There are currently two main methods for pushing access control advertisements:

[0003] Static display: Fixed advertising display boards, such as paper posters and lightbox advertisements, are set up in the access control area. The content of these advertisements is pre-designed and remains unchanged for a long time, displaying the same advertising information to all people passing through the access control.

[0004] Screen playback: An LCD screen is installed to play preset advertising videos or images in a loop. The playback order and content are also set in advance, without taking into account the specific circumstances of the people passing by.

[0005] However, the above method has the following application drawbacks:

[0006] Static display and screen-based advertising methods cannot personalize ads based on different user characteristics and behaviors. Regardless of a user's location, movement speed, or orientation, they will only see the same ad content, failing to meet the needs and interests of different users, thus significantly reducing the attractiveness and effectiveness of the ads.

[0007] The current approach is essentially a one-way information dissemination, where users can only passively receive advertisements and cannot effectively interact with them. Users cannot actively select advertising content that interests them or perform related actions, resulting in low user engagement and difficulty in forming a lasting impression.

[0008] Static display boards and traditional screen advertisements are two-dimensional, lacking a sense of depth and realism, making it difficult to stand out among a sea of ​​information and attract users' attention. Summary of the Invention

[0009] To address the technical problems existing in the prior art, the present invention provides the following technical solution:

[0010] On the one hand, a method for pushing access control advertisements based on 3D imaging is provided. This method is implemented by an electronic device and includes:

[0011] S1, 3D Imaging Data Acquisition: A 3D depth sensor is used to acquire depth images and RGB images within the access control area, and the depth images are converted into 3D point clouds in the camera coordinate system;

[0012] S2. User Detection and Tracking: The 3D point cloud is preprocessed, and the user point cloud is segmented using a region growing algorithm to extract the position information of the user center; the Kalman filter algorithm is used to track the user's position and velocity, and the user's orientation angle is calculated.

[0013] S3. Ad content adaptation: Based on the user's position, speed, and orientation angle, adjust the translation, rotation, and scaling parameters of the 3D advertising model so that the 3D advertising model is located in the optimal field of view in front of the user.

[0014] S4, 3D Holographic Display: Project the adapted 3D model of the advertisement onto a holographic display to display it as a free-space 3D advertisement;

[0015] S5. Interactive Response: Recognizes user gestures through a 3D depth sensor or receives user commands through AI glasses to respond to changes in advertising content or redirection of links.

[0016] Preferably, the formula for converting the depth image into a 3D point cloud in step (1) is:

[0017]

[0018] Where (u, v) are the pixel coordinates of the depth image, and (d) is the depth value corresponding to that pixel. , () represents the coordinates of the camera's principal point. , (x, y, z) represents the camera focal length, and (x, y, z) represents the 3D point coordinates in the camera coordinate system.

[0019] Preferably, in step (2), the seed point of the region growing algorithm is selected as the point closest to the camera in the depth image, and the growing conditions include:

[0020] (a) Normal angle condition: The cosine of the normal angle between the neighboring point and the seed point is greater than 0.9;

[0021] (b) Distance condition: The Euclidean distance between the neighboring point and the seed point is less than 50 mm.

[0022] Preferably, the state vector of the Kalman filter in step (2) is:

[0023] ,

[0024] in:

[0025] (X,Y,Z) represents the camera coordinates at the user center.

[0026] For user movement speed;

[0027] ΔX, ΔY, and ΔZ are the components of the user's movement velocity in the X, Y, and Z axis directions, respectively (unit: mm / s).

[0028] in The sampling time, Δt, is in seconds (s) and is used to control the time resolution of the Kalman filter.

[0029] Preferably, the translation transformation of the 3D advertising model in step (3) is as follows:

[0030] Translate the model center to ,

[0031] in (T) represents the camera coordinates of the user center at time t (in mm), and (D) represents the distance between the user and the advertisement (taken as 1500 mm).

[0032] Rotation transformation is a rotation about the Y-axis with the user's orientation angle. The rotation matrix is:

[0033]

[0034] in The unit is radians (rad), and the angle between the principal direction of the user point cloud and the X-axis of the camera coordinate system is calculated using the arctangent function.

[0035] Preferably, the gesture recognition in step (5) uses the PointNet++ model, which includes hand point cloud segmentation, hierarchical feature extraction, and gesture classification. The classification loss function is cross-entropy.

[0036]

[0037] in For gesture category labels, The class probabilities predicted by the model;

[0038] Where N is the total number of samples (i.e. the number of gesture samples participating in the classification).

[0039] On the other hand, a 3D imaging-based access control advertising push device is provided, the device comprising:

[0040] (1) 3D depth sensor: used to acquire depth images and RGB images within the access control area;

[0041] (2) Edge computing terminal: used to execute the above-described access control advertising push method based on 3D imaging, including 3D point cloud conversion, user detection and tracking, and advertising content adaptation;

[0042] (3) 3D holographic display: used to display 3D advertisements in free space;

[0043] (4) AI Interactive Terminal: Used to receive user interaction instructions.

[0044] Preferably, the 3D depth sensor is a structured light camera or a TOF camera, installed on the wall above the access control system in the elevator / stairwell, covering the area that the user must pass through when opening the door;

[0045] The 3D holographic display uses a spatial light modulator (SLM) to project the 3D model of the advertisement into a free-space holographic image;

[0046] The AI ​​interactive terminal includes AI glasses, which are used to overlay augmented reality advertisements and receive user click commands.

[0047] On the other hand, an electronic device is provided, comprising: a processor; and a memory storing computer-readable instructions, which, when executed by the processor, implement any of the methods described above for push-to-access-advertisements-based-3D-imaging-based-access-control-systems-as-described-methods-for-3D-imaging-based-advertisements-as-described-methods-for-3D-imaging-based-access-control-advertisements ...as-described-methods-for-3D-imaging-as-described-

[0048] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, the at least one instruction being loaded and executed by a processor to implement any of the above-described methods for pushing access control advertisements based on 3D imaging.

[0049] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0050] This invention, through 3D imaging data acquisition, user detection, and tracking, can accurately obtain information such as the user's position, speed, and orientation angle. Based on this information, the 3D model of the advertisement is translated, rotated, and scaled to ensure that the advertisement is always located in the optimal field of vision in front of the user. This enables personalized advertising based on the user's specific situation, improving the targeting and attractiveness of the advertisement.

[0051] This invention incorporates an interactive response mechanism, allowing users to switch between advertising content or navigate to new links via gestures or AI glasses. This interactive approach enables users to actively participate in the advertising experience, increasing user engagement and attention to the ads, thus enhancing the effectiveness of ad dissemination.

[0052] This invention employs 3D holographic display technology to project the adapted 3D advertising model onto a holographic display, displaying it as a free-space 3D advertisement. Compared to traditional two-dimensional flat advertisements, 3D holographic advertisements have a stronger sense of three-dimensionality and realism, bringing users a brand-new visual experience, more easily attracting users' attention, and improving advertisement memorability. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart of an access control advertising push method based on 3D imaging provided by an embodiment of the present invention;

[0055] Figure 2 This is a logical diagram illustrating the core steps provided in the embodiments of the present invention;

[0056] Figure 3 This is a block diagram of an access control advertising push device based on 3D imaging provided in an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0058] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0059] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0060] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0061] In this embodiment of the invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0062] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0063] This invention provides a method for pushing advertisements to access control systems based on 3D imaging. This method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart shown is for a 3D imaging-based access control advertising push method. The processing flow of this method may include the following steps:

[0064] S1, 3D Imaging Data Acquisition: A 3D depth sensor is used to acquire depth images and RGB images within the access control area, and the depth images are converted into 3D point clouds in the camera coordinate system;

[0065] S2. User Detection and Tracking: The 3D point cloud is preprocessed, and the user point cloud is segmented using a region growing algorithm to extract the position information of the user center; the Kalman filter algorithm is used to track the user's position and velocity, and the user's orientation angle is calculated.

[0066] S3. Ad content adaptation: Based on the user's position, speed, and orientation angle, adjust the translation, rotation, and scaling parameters of the 3D advertising model so that the 3D advertising model is located in the optimal field of view in front of the user.

[0067] S4, 3D Holographic Display: Project the adapted 3D model of the advertisement onto a holographic display to display it as a free-space 3D advertisement;

[0068] S5. Interactive Response: Recognizes user gestures through a 3D depth sensor or receives user commands through AI glasses to respond to changes in advertising content or redirection of links.

[0069] This invention utilizes 3D imaging technology, such as 3D imaging in access control (or stairwells, elevators), to display different access control advertisements and even interact with users. The solution is novel and forcibly displays advertisements to users. (3D imaging can also interact with AI glasses to conduct commercial and advertising link interactions.)

[0070] like Figure 2 As shown, the core process of this method is: 3D imaging data acquisition → point cloud preprocessing → user detection and tracking → advertising content adaptation → 3D holographic display → interactive response. The system consists of the following modules: 3D depth sensor (data acquisition), edge computing terminal (data processing and algorithm execution), 3D holographic display (advertisement display), and AI interactive terminal (such as AI glasses, optional). System workflow summary.

[0071] When a user opens the door, the 3D depth camera is activated, capturing depth and RGB images of the area where the user is located.

[0072] The edge computing terminal converts the depth image into a 3D point cloud and performs preprocessing (denoising).

[0073] The user point cloud is segmented using a region growing algorithm to extract the user's center location;

[0074] Use Kalman filtering to track the user's position and velocity, and calculate the user's orientation;

[0075] Adjust the translation, rotation, and scaling parameters of the 3D advertising model based on the user's location and orientation;

[0076] The adapted advertising model is projected onto a holographic display and displayed as a free-space 3D advertisement.

[0077] Users interact with advertisements through AI glasses or gestures (such as clicking links or switching content), and the system responds to the interaction commands (such as jumping to an e-commerce platform or switching advertisements).

[0078] The technical application principle of the present invention will be described in detail below.

[0079] 1. 3D Imaging Data Acquisition and Preprocessing

[0080] Equipment selection: Use a structured light depth camera (such as Apple TrueDepth) or a TOF (Time-of-Flight) camera, installed on the wall above the access control system in the elevator / stairwell, covering the area that users must pass through when opening the door (field of view: 120° horizontal, 80° vertical, depth range 0.5-5m).

[0081] Data format conversion: The depth camera outputs a depth image (pixel values ​​are depth values ​​(d), in mm) and an RGB image (pixel values ​​are color information). The formula for converting the depth image to a 3D point cloud in the camera coordinate system is as follows:

[0082]

[0083] in:

[0084] (u, v): Pixel coordinates (column, row) of the depth image;

[0085] (d): The depth value (mm) corresponding to this pixel;

[0086] : Camera principal point coordinates (in pixels, obtained from camera calibration, such as those of a 1920×1080 resolution camera). , );

[0087] Camera focal length (in pixels, obtained from camera calibration, such as...) );

[0088] (X, Y, Z): 3D point coordinates in the camera coordinate system (mm, (Z) axis is the direction of the camera optical axis).

[0089] Point cloud preprocessing: Statistical filtering is used to remove outliers (such as clutter in the environment). The steps are as follows:

[0090] For each point Take its (k) neighbor points ( (experience points);

[0091] Calculate the neighboring points and average distance ;

[0092] like ( The mean of the average distances between all points. If the standard deviation is 0, then delete. .

[0093] 2. User Detection and Tracking

[0094] User detection: A region growing algorithm is used to segment user point clouds from the preprocessed point cloud. The principle is to grow regions based on the geometric continuity (normal direction and distance) of the point cloud. The steps are as follows:

[0095] Seed point selection: Select the point in the depth image that is closest to the camera (with the smallest (Z) value) as the initial seed point. ;

[0096] Regional growth conditions: for the (k) neighboring points of the seed point If the following conditions are met, then Join the user area:

[0097] Conditions for the included angle of the normals: ( For the seed point normal, For the normal of the neighboring points, ,correspond );

[0098] Distance conditions: ( (Experience points).

[0099] Normal calculation: for each point (k) Neighborhood Point Set Principal component analysis (PCA) is used to calculate the covariance matrix. And solve for the eigenvalues ​​and eigenvectors:

[0100]

[0101] in:

[0102] , where is the mean of the neighborhood points.

[0103] eigenvalues ​​of the covariance matrix Minimum eigenvalue The corresponding feature vector is a point. normal direction The symbol T represents the matrix transpose operator, used to convert a column vector into a row vector. In the calculation of the covariance matrix, the rows and columns of the vectors are interchanged by transposing the deviation vectors of the neighborhood points from the mean (i.e., marked with "T" in the upper right corner), thus obtaining the covariance matrix through matrix multiplication.

[0104] The specific steps are as follows: First, calculate the mean of the neighborhood points and construct the covariance matrix to describe the spatial distribution characteristics of the point set; then, perform eigenvalue decomposition on the covariance matrix to obtain three eigenvalues. The eigenvector corresponding to the smallest eigenvalue λ3 is the normal direction of that point.

[0105] The meaning of eigenvalues ​​in the covariance matrix: Eigenvalues ​​reflect the degree of dispersion of the point set along the corresponding eigenvector direction (data variance). The largest eigenvalue λ1 corresponds to the direction of the most dispersed point cloud distribution (main direction), and the smallest eigenvalue λ3 corresponds to the direction of the most concentrated distribution. Its eigenvector is perpendicular to the point cloud surface, so it is used as the normal direction to determine whether the neighboring points satisfy the geometric continuity condition of region growth (normal angle threshold), thereby realizing the segmentation of the user point cloud from the environmental background.

[0106] User tracking: A Kalman filter is used to predict the user's position and velocity in the camera coordinate system to achieve continuous tracking. The state vector is defined as follows:

[0107] ,

[0108] Where: T represents the transpose operation of a matrix or vector, used to convert a column vector into a row vector (or vice versa).

[0109] Where ((X,Y,Z)) represents the camera coordinates at the user center. User movement speed (unit: mm / s).

[0110] State transition equation (prediction steps):

[0111]

[0112] in:

[0113] The state transition matrix ( (The sampling time is taken as 0.1s), for example, the state transition matrix. for:

[0114]

[0115] In the matrix, "0.1" represents the specific value of the sampling time Δt, corresponding to the parameter setting "Δt=0.1s" in the document, which is used to describe the relationship between the position component and the velocity and time.

[0116] The process noise (with a mean of 0 and a covariance of ) Gaussian distribution, Take a diagonal matrix with diagonal elements ([100,100,100,10,10,10])).

[0117] : Predict the state vector, which includes the user's position (X,Y,Z) and velocity (ΔX,ΔY,ΔZ);

[0118] Observation equations (update steps):

[0119]

[0120] in:

[0121] These are the observation values ​​(camera coordinates at the user center, output by the region growing algorithm);

[0122] For the observation matrix (extracting positional information from the state vector):

[0123] ,

[0124] The observed noise (following a mean of 0 and a covariance of ) Gaussian distribution, Take a diagonal matrix with diagonal elements ([100,100,100])). It follows a Gaussian distribution with a mean of 0, and its covariance matrix R is a 3×3 diagonal matrix. The diagonal elements [ [100,100,100] ] correspond to the noise variance (unit: mm²) in the X, Y, and Z axes, respectively, and are used to quantify the uncertainty of the observations.

[0125] Kalman filter iterative steps:

[0126] Prediction: Calculate the predicted state Covariance of prediction:

[0127] ;

[0128] in To predict the covariance matrix, Let be the covariance matrix of the previous time step. This is the transpose of the state transition matrix. Q is the process noise covariance matrix, a 6×6 diagonal matrix. The diagonal elements ([100,100,100,10,10,10]) correspond to the noise variance (unit: mm² / s²) of the user's position (X, Y, Z axes) and velocity (ΔX, ΔY, ΔZ axes), respectively, and are used to quantify the uncertainty of process noise in the system model.

[0129] Calculate the Kalman gain:

[0130] ;

[0131] in Here is the Kalman gain matrix. For the transpose of the observation matrix, This represents the matrix inversion operation; R is the observation noise covariance matrix, a 3×3 diagonal matrix, where the diagonal elements ([100,100,100]) correspond to the observation noise variance (unit: mm²) of the user-center camera coordinates (X, Y, Z axes), used to describe the error characteristics of the observations output by the region growing algorithm;

[0132] Update: Correct the predicted state using the observed values ​​to obtain the current state:

[0133] ,

[0134] And update the covariance:

[0135] ( (The identity matrix).

[0136] in The updated covariance matrix, It is a 6×6 identity matrix.

[0137] 3. Ad content adaptation algorithm

[0138] Adaptation logic: Based on the user's location ( ) and posture (orientation) (Calculated from the main direction of the user's point cloud), adjust the display position, rotation angle and scaling ratio of the 3D model of the advertisement so that the advertisement is always located in the optimal field of view in front of the user (such as 1.5m directly in front of the user, at the same height as the user's eyes).

[0139] Advertising model transformation:

[0140] Translation transformation: Translate the center of the advertising model from the origin of the world coordinate system to the target position in front of the user. ( mm (distance between user and advertisement);

[0141] Rotation Transformation: Rotate the advertising model around the Y-axis (User orientation angle, derived from the principal direction vector of the user point cloud) calculate, ), rotation matrix for:

[0142] in The component of the main direction vector of the user point cloud on the X-axis (unit: mm). The component in the Z-axis direction (unit: mm).

[0143] ,

[0144] Scaling transformation: based on the distance between the user and the ad. Adjust ad size and scaling factor The calculation is as follows:

[0145] in:( (Advertisement size at a baseline distance (e.g., 2m)).

[0146] in The advertising size parameters are based on the reference distance (unit: mm, such as 2000 mm).

[0147] Projection transformation: The 3D model of the advertisement is transformed from the world coordinate system to the camera coordinate system, and then projected onto the screen coordinate system (the pixel coordinates of the holographic display). The formula is:

[0148]

[0149] in:

[0150] The world coordinates for the advertising model;

[0151] The camera intrinsic parameter matrix (by...) composition):

[0152] ,

[0153] Screen pixel coordinates ( , , (The coordinates of the advertising model are in the camera coordinate system).

[0154] 4. 3D Holographic Display and Interaction Design

[0155] Display device: A holographic display based on a spatial light modulator (SLM) projects the adapted 3D advertising model into a free-space holographic image (users can directly observe the 3D advertisement suspended in the air without wearing glasses).

[0156] Interaction methods: Supports both AI glasses interaction and gesture interaction modes.

[0157] AI Glasses Interaction: Users wear AI glasses that integrate depth sensors and display modules. The glasses receive advertising content sent by the access control system via wireless communication (such as Wi-Fi 6), and augmented reality (AR) advertisements are overlaid on the glasses' display interface (such as clicking on a "coupon" link in the advertisement, which directly redirects to an e-commerce platform).

[0158] Gesture interaction: The 3D depth camera recognizes user gestures (such as "click" and "swipe") to control the switching of advertising content (such as swiping to switch to the next advertisement, and clicking to open the advertisement link).

[0159] Gesture recognition algorithm: 3D gesture classification is implemented using PointNet++ (a point cloud deep learning model). The steps are as follows:

[0160] Hand point cloud segmentation: Extracting the hand region from the user's point cloud (depth threshold: , (depth value at the center of the user's body).

[0161] Where Z is the depth value of the current point (unit: mm), and Z_{body} is the depth value of the user's body center (unit: mm, extracted from the user's point cloud center coordinates).

[0162] Feature extraction: Local and global features of the hand point cloud (such as finger joint positions and palm shape) are extracted using PointNet++'s Set Abstraction module.

[0163] Gesture classification: Features are input into a fully connected layer and a softmax function, and the output is a gesture category (e.g., "tap", "swipe", "clench fist"). The classification loss function is cross-entropy.

[0164]

[0165] in For gesture category labels (one-hot encoding) This represents the class probabilities predicted by the model.

[0166] Where N is the total number of gesture categories (e.g., 3 categories: tap, swipe, fist).

[0167] Example: Intelligent Elevator Advertising Delivery System

[0168] A 3D intelligent advertising system was deployed in the office building elevator to optimize the advertising display effect by dynamically adapting to the user's perspective. The elevator car measures 2.5m (height) × 1.6m (width) × 1.8m (depth). The advertising screen is installed on the rear wall of the car (1.2m × 0.8m), and an Azure Kinect DK 3D depth sensor is installed in the center of the top.

[0169] S1: 3D Imaging Data Acquisition

[0170] Sensor parameters: depth image resolution 640×576, RGB image resolution 1920×1080, acquisition frame rate 30fps, depth detection range 0.5-5m (effective distance 0.5-2m inside elevator).

[0171] Data output: Real-time generation of 3D point cloud in camera coordinate system (point cloud density 1000 points / ㎡), where the origin of the camera coordinate system is the sensor optical center (x-axis horizontal to the right, y-axis vertical downward, z-axis pointing towards the car door).

[0172] S2: User Detection and Tracking

[0173] Preprocessing: Voxel filtering (voxel size 0.01m) and outlier removal (points with fewer than 5 neighbors within a radius of 0.1m are considered noise) are performed on the original point cloud.

[0174] Region growing algorithm: The seed point is set as the point cloud cluster that is closest to the sensor in the z-axis direction (car depth direction) (threshold: points with a distance of <0.2m from the seed point are included in the same region), and the user point cloud is segmented (assuming a single user scenario, the number of point clouds is about 8000-12000 points).

[0175] User center location: Extract the geometric center of the user point cloud, example coordinates (x=0.3m, y=0.0m, z=1.5m) (that is, the user is located 1.5m to the right front of the sensor).

[0176] Kalman filter tracking:

[0177] State vector: [position x, position y, position z, velocity vx, velocity vy, velocity vz], initial velocity is set to 0.1m / s (walking speed inside the elevator).

[0178] The process noise covariance Q = diag([0.01, 0.01, 0.01, 0.001, 0.001, 0.001]), and the measurement noise covariance R = diag([0.05, 0.05, 0.05, 0.01, 0.01, 0.01]).

[0179] Orientation angle calculation: The orientation angle θ is the angle between the body axis direction extracted by principal component analysis (PCA) of the user's point cloud and the normal direction (z-axis) of the advertising screen. Example: θ = 0° when the user is facing the advertising screen directly, and θ = 30° when the user is turned to the side.

[0180] S3: Ad Content Adaptation

[0181] Initial parameters of the 3D model of the advertisement: size 1.2m×0.8m×0.02m (width×height×thickness), initial position (0, 0,0) (center of the advertisement screen), rotation angle 0°, scaling ratio 1.0.

[0182] Dynamic adjustment rules:

[0183] Translation: Adjust the model's x-axis position based on the user's x-coordinate offset Δx. Formula: Model x-coordinate = User x-coordinate × 0.3 (proportional coefficient to avoid excessive offset). Example: When user x = 0.3m, model x = 0.09m (fine-tuning towards the user).

[0184] Rotation: Adjust the model's rotation angle around the y-axis according to the orientation angle θ. Formula: Rotation angle = -θ (opposite to the user's orientation). Example: User θ = 30° (facing right), model rotates -30° (rotates to the left, facing the user).

[0185] Scaling: Adjust the scaling ratio based on the user's z-coordinate (distance). Formula: Scaling ratio = 1.0 + (2.0m - z) × 0.2 (the closer the distance, the larger the scaling, up to a maximum of 1.2 times). Example: When the user's z = 1.5m, the scaling ratio = 1.0 + (2.0 - 1.5) × 0.2 = 1.1.

[0186] Once a user enters the elevator, the system completes detection and parameter adjustment within 0.3 seconds. The 3D advertising model automatically rotates and slightly shifts to ensure that the user can see the advertising content clearly and directly, regardless of their standing position or orientation. For example, if a user is standing on the left side of the elevator (x=-0.5m) and facing the elevator door (θ=180°), the model will shift 0.15m to the left and rotate 180° to ensure the advertisement faces the user directly.

[0187] like Figure 3 As shown, on the other hand, a 3D imaging-based access control advertising push device is provided, the device comprising:

[0188] (1) 3D depth sensor: used to acquire depth images and RGB images within the access control area;

[0189] (2) Edge computing terminal: used to execute the above-described access control advertising push method based on 3D imaging, including 3D point cloud conversion, user detection and tracking, and advertising content adaptation;

[0190] (3) 3D holographic display: used to display 3D advertisements in free space;

[0191] (5) AI Interactive Terminal: Used to receive user interaction instructions.

[0192] Preferably, the 3D depth sensor is a structured light camera or a TOF camera, installed on the wall above the access control system in the elevator / stairwell, covering the area that the user must pass through when opening the door;

[0193] The 3D holographic display uses a spatial light modulator (SLM) to project the 3D model of the advertisement into a free-space holographic image;

[0194] The AI ​​interactive terminal includes AI glasses, which are used to overlay augmented reality advertisements and receive user click commands.

[0195] The composition and interaction of the above-mentioned devices can be understood by referring to the principles of the methods described above.

[0196] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 4 As shown, the electronic device may include the above-mentioned Figure 3 The illustrated access control advertising push device is based on 3D imaging. Optionally, the electronic device 410 may include a first processor 2001.

[0197] Optionally, the electronic device 410 may also include a memory 2002 and a transceiver 2003.

[0198] The first processor 2001, memory 2002, and transceiver 2003 can be connected via a communication bus.

[0199] The following is combined Figure 4 A detailed description of each component of electronic device 410 is provided below:

[0200] The first processor 2001 is the control center of the electronic device 410. It can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0201] Optionally, the first processor 2001 can perform various functions of the electronic device 410 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0202] In a specific implementation, as one example, the first processor 2001 may include one or more CPUs, for example... Figure 4 CPU0 and CPU1 are shown in the diagram.

[0203] In a specific implementation, as one example, the electronic device 410 may also include multiple processors, for example... Figure 4 The first processor 2001 and the second processor 2004 are shown in the diagram. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). Here, a processor can refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0204] The memory 2002 is used to store the software program that executes the present invention, and is controlled by the first processor 2001 to execute it. The specific implementation method can be referred to the above method embodiment, and will not be repeated here.

[0205] Optionally, the memory 2002 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 2002 may be integrated with the first processor 2001 or may exist independently and be connected via the interface circuit of the electronic device 410. Figure 4 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0206] The transceiver 2003 is used to communicate with network devices or with terminal devices.

[0207] Alternatively, transceiver 2003 may include a receiver and a transmitter. Figure 4 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function.

[0208] Optionally, the transceiver 2003 can be integrated with the first processor 2001, or it can exist independently and be connected via the interface circuit of the electronic device 410. Figure 4 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0209] It should be noted that, Figure 4 The structure of the electronic device 410 shown does not constitute a limitation on the router. Actual knowledge structure identification devices may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0210] Furthermore, the technical effects of the electronic device 410 can be referred to the technical effects of the access control advertising push method based on 3D imaging described in the above method embodiments, and will not be repeated here.

[0211] It should be understood that the first processor 2001 in the embodiments of the present invention may be a central processing unit (CPU), or it may 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 any conventional processor, etc.

[0212] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0213] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0214] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0215] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0216] It should be understood that, in various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0217] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0218] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0219] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0220] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0221] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0222] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0223] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for pushing access control advertisements based on 3D imaging, characterized in that, The method includes: S1, 3D Imaging Data Acquisition: A 3D depth sensor is used to acquire depth images and RGB images within the access control area, and the depth images are converted into 3D point clouds in the camera coordinate system; S2. User Detection and Tracking: The 3D point cloud is preprocessed, and the user point cloud is segmented using a region growing algorithm to extract the position information of the user center; the Kalman filter algorithm is used to track the user's position and velocity, and the user's orientation angle is calculated. S3. Ad content adaptation: Based on the user's position, speed, and orientation angle, adjust the translation, rotation, and scaling parameters of the 3D advertising model so that the 3D advertising model is located in the optimal field of view in front of the user. S4, 3D Holographic Display: Project the adapted 3D model of the advertisement onto a holographic display to display it as a free-space 3D advertisement; S5. Interactive Response: Recognizes user gestures through a 3D depth sensor or receives user commands through AI glasses to respond to changes in advertising content or redirection of links.

2. The access control advertising push method based on 3D imaging according to claim 1, characterized in that, The formula for converting the depth image into a 3D point cloud in step (1) is: ， Where (u, v) are the pixel coordinates of the depth image, and (d) is the depth value corresponding to that pixel. , () represents the coordinates of the camera's principal point. , (x, y, z) represents the camera focal length, and (x, y, z) represents the 3D point coordinates in the camera coordinate system.

3. The access control advertising push method based on 3D imaging according to claim 1, characterized in that, In step (2), the seed point of the region growing algorithm is selected as the point closest to the camera in the depth image, and the growing conditions include: (a) Normal angle condition: The cosine of the normal angle between the neighboring point and the seed point is greater than 0.9; (b) Distance condition: The Euclidean distance between the neighboring point and the seed point is less than 50 mm.

4. The access control advertising push method based on 3D imaging according to claim 1, characterized in that, The state vector of the Kalman filter in step (2) is: ， in: (X,Y,Z) represents the camera coordinates at the user center. For user movement speed; ΔX, ΔY, and ΔZ are the components of the user's movement velocity in the X, Y, and Z axis directions, respectively (unit: mm / s). in The sampling time, Δt, is in seconds (s) and is used to control the time resolution of the Kalman filter.

5. The access control advertising push method based on 3D imaging according to claim 1, characterized in that, The translation transformation of the 3D advertising model in step (3) is as follows: Translate the model center to , in (T) represents the camera coordinates of the user center at time t (in mm), and (D) represents the distance between the user and the advertisement (taken as 1500 mm). Rotation transformation is a rotation about the Y-axis to the user's orientation angle. The rotation matrix is: ， in The unit is radians (rad), and the angle between the principal direction of the user point cloud and the X-axis of the camera coordinate system is calculated using the arctangent function.

6. The access control advertising push method based on 3D imaging according to claim 1, characterized in that, The gesture recognition in step (5) uses the PointNet++ model, which includes hand point cloud segmentation, hierarchical feature extraction, and gesture classification. The classification loss function is cross-entropy. ， in For gesture category labels, The predicted class probabilities by the model; Where N is the total number of samples (i.e. the number of gesture samples participating in the classification).

7. A 3D imaging-based access control advertising push device, characterized in that, The device includes: (1) 3D depth sensor: used to acquire depth images and RGB images within the access control area; (2) Edge computing terminal: used to execute the access control advertising push method based on 3D imaging as described in any one of claims 1-6, including 3D point cloud conversion, user detection and tracking, and advertising content adaptation; (3) 3D holographic display: used to display 3D advertisements in free space; (4) AI Interactive Terminal: Used to receive user interaction instructions.

8. The access control advertising push device based on 3D imaging according to claim 7, characterized in that, The 3D depth sensor is a structured light camera or a TOF camera, installed on the wall above the access control system in the elevator / stairwell, covering the area that users must pass through when opening the door; The 3D holographic display uses a spatial light modulator (SLM) to project the 3D model of the advertisement into a free-space holographic image; The AI ​​interactive terminal includes AI glasses, which are used to overlay augmented reality advertisements and receive user click commands.

9. An electronic device, characterized in that, The electronic device includes: processor; A memory storing computer-readable instructions that, when executed by the processor, implement the method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains program code that can be invoked by a processor to execute the method as described in any one of claims 1 to 6.

Images