Advertisement signboard playing method based on passerby vision and advertisement signboard

Abstract

The invention discloses an advertisement signboard playing method based on passerby vision and an advertisement signboard, and relates to the field of image processing. The method is applied to an edge server, the edge server is in communication connection with one or more advertisement signboards in a target area, and the method comprises the following steps: obtaining target images of passersby in the area range of the advertisement signboards; obtaining visual features and group features of pedestrians in the target image, and determining one or more target advertisements based on the visual features and the group features; obtaining a track of a pedestrian in the target image; and controlling one or more advertisement signboards to play each target advertisement based on the track. By implementing the technical scheme provided by the invention, the advertisement putting effect can be improved.

Description

Technical Field

[0001] The present application relates to the field of image processing, and in particular to an advertising signboard playback method based on passerby vision and an advertising signboard. Background Art

[0002] With the advancement of smart city construction, various types of advertising signs are widely used in public places such as commercial centers and transportation hubs. These advertising signs usually use a fixed playback method to display advertising content to different groups of passers-by. However, different groups of passers-by often show different behavior patterns and visual characteristics when watching advertisements, such as viewing angles, dwelling time, group density, etc. How to improve the pertinence and effectiveness of advertising sign placement has become an important issue that needs to be solved in the field of advertising placement.

[0003] At present, the main way of playing advertising signs is to loop different advertising content in a preset time period, or switch advertising content based on basic characteristics of passers-by such as age and gender based on face recognition technology. For example, some advertising sign systems will play workplace-related advertising content during rush hour and parent-child activities-related advertising content on weekends.

[0004] However, existing technologies only deliver advertising content based on time periods or single demographic characteristics, and fail to accurately grasp the variability of passerby groups. This results in a mismatch between advertising content and actual viewers, and inappropriate advertising timing, resulting in low advertising delivery effects. Summary of the invention

[0005] The present application provides an advertising signboard playback method and an advertising signboard based on passerby vision, which can improve the effect of advertising delivery.

[0006] In a first aspect of the present application, a method for playing an advertising sign based on passerby vision is provided, which is applied to an edge server, wherein the edge server is communicatively connected to one or more advertising signs in a target area, and the method for playing an advertising sign based on passerby vision includes: Acquire a target image of a passerby within the area of ​​the advertising signboard; Acquire visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; Obtaining the trajectory of the passerby in the target image; Based on the travel trajectory, one or more of the advertising signs are controlled to play the target advertisements.

[0007] In a second aspect of the present application, an edge server is provided, including: Acquire target images of passers-by within the advertising sign area; Acquire visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; Obtaining the trajectory of the passerby in the target image; Based on the travel trajectory, one or more of the advertising signs are controlled to play the target advertisements.

[0008] In a third aspect of the present application, an advertising signboard is provided, characterized in that it includes: Acquire target images of passers-by within the area; Acquire visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; Obtaining the trajectory of the passerby in the target image; Based on the travel trajectory, one or more other advertising signs are controlled to play each of the target advertisements.

[0009] In a fourth aspect of the present application, a computer storage medium is provided, wherein the computer storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above method steps.

[0010] In this application of A fifth aspect provides an electronic device, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0011] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: By acquiring the target image of passers-by in the area of ​​the billboard, extracting the visual features and group features of the passers-by from the target image, determining the target advertisement based on these features, and obtaining the pedestrian's track to control the playback of the billboard, the real-time perception and response to the changes in the pedestrian group is achieved. Since the edge server is connected to the billboard in the target area, the system can quickly process the target image and adjust the advertising playback strategy in time. By analyzing the visual features and group features of the passers-by to determine the target advertisement, and combining the track to control the playback timing, the limitation of relying solely on the preset time period or the characteristics of a single group of people for advertising is avoided, and the matching degree between the advertising content and the actual viewers is improved. This solution can not only accurately grasp the variability of the pedestrian group, but also select the appropriate playback time, which can improve the effect of advertising. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1It is a flow chart of a method for playing advertising signs based on passerby vision provided in an embodiment of the present application; Figure 2 It is a flow chart of another method for playing advertising signs based on passerby vision provided in an embodiment of the present application; Figure 3 It is a structural diagram of an edge server provided in an embodiment of the present application; Figure 4 It is a structural schematic diagram of an advertising sign provided in an embodiment of the present application; Figure 5 It is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0013] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0014] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0015] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0016] This application provides a method for playing advertising signs based on passerby vision, referring to Figure 1 , Figure 1 This is a flow chart of a method for playing advertising signs based on passerby vision disclosed in an embodiment of the present application. The method is applied to an edge server, which is connected to one or more advertising signs in a target area. The specific steps are as follows: Step 101: Acquire a target image of a passerby within the area of ​​the advertising sign.

[0017] The target area refers to a specific spatial area formed by the effective coverage of one or more advertising signs. Specifically, the target area can be an open area inside a public place such as a shopping mall, airport, subway station, etc., where at least one advertising sign is set up and passers-by can move freely and view advertising content.

[0018] In the embodiment of the present application, the target area can be understood as a collection of spatial ranges formed by extending outward a specific distance with each advertising sign as the center. Among them, the effective coverage range of each advertising sign can be a circular area, and the radius of the circular area can be determined according to factors such as the size of the advertising sign, the installation height, and the clarity of the broadcast content. For example, for a standard LED advertising screen of 2 meters by 3 meters, its effective coverage range can be set to a circular area with the center point of the advertising screen as the center and a radius of 10 meters. When the effective coverage ranges of multiple advertising signs overlap, these overlapping areas are also part of the target area.

[0019] The division of the target area is used to determine the spatial range that the edge server needs to monitor and analyze, so that the system can accurately obtain the visual behavior and group characteristics of passers-by within the range. By reasonably dividing the target area, it can ensure that the system can effectively capture the interaction between passers-by and advertising signs, while avoiding processing regional information that is not related to advertising placement, thereby improving the operating efficiency of the system.

[0020] Correspondingly, the advertising sign area range refers to the effective broadcast coverage range of a single advertising sign, which is the spatial area where passers-by can clearly see the advertising content and have effective visual interaction. This area is different from the aforementioned target area. The target area is a collection of the coverage ranges of multiple advertising signs, while the advertising sign area range is the specific scope defined for a single advertising sign.

[0021] In the embodiment of the present application, the area range of the advertising signboard can be understood as a fan-shaped space centered on the advertising signboard. The size and shape of the fan-shaped space are jointly determined by multiple parameters: its radial distance is determined by the size, installation height and display resolution of the advertising signboard, and is generally set to the maximum distance that allows passers-by to clearly identify the advertising content; its opening angle is determined by the display screen viewing angle of the advertising signboard and the environmental occlusion, usually 120-160 degrees; its height range takes into account the line of sight of ordinary passers-by, which is about 1.2 meters to 2 meters from the ground. For example, for a 55-inch high-definition LED advertising screen, its area range may be a fan-shaped area with the advertising screen as the vertex, a radius of 8 meters, and an opening angle of 140 degrees.

[0022] Specifically, three high-definition cameras can be deployed within the area of ​​each advertising sign, and they are set at different positions to achieve all-round monitoring. Among them, the first camera is installed directly above the advertising sign to take a bird's-eye view of the distribution of people in the entire fan-shaped area; the second and third cameras are installed on the left and right sides of the advertising sign, respectively, at a 45-degree angle to the surface of the sign, to collect side images of passers-by. This three-point arrangement can effectively avoid blind spots and ensure that the system can accurately capture the complete behavioral characteristics of passers-by within the area of ​​the advertising sign. The installation height of all cameras is uniformly 3 meters from the ground. This height can ensure a larger monitoring range without causing obvious interference to passers-by.

[0023] After the edge server receives the raw images transmitted by multiple cameras, it first performs image preprocessing. The preprocessing process includes operations such as denoising, illumination equalization, and geometric correction to eliminate the impact of environmental factors on image quality. Next, the system synchronizes the image data from different cameras in time and aligns them in space. Time synchronization ensures that the images collected by different cameras correspond to the same moment, and spatial alignment establishes the correspondence between images from different perspectives. Through feature point matching and image stitching algorithms, the system fuses the image data from three perspectives into a complete set of target images.

[0024] In order to adapt to the needs of different scenarios, the system adopts an adaptive image acquisition strategy. In standard mode, each camera acquires images at a frame rate of 25 frames per second, and the system processes a set of target images every 200 milliseconds. When a sudden increase in the flow of people or a gathering of groups is detected, the system will automatically increase the sampling frequency to a set every 100 milliseconds to ensure that important behavioral features are not missed. Conversely, during periods of sparse flow of people, the system will appropriately reduce the sampling frequency to save computing resources.

[0025] Step 102: Obtain visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and group features.

[0026] Among them, visual features refer to the behavioral characteristics related to visual perception exhibited by passers-by within the area of ​​the advertising signboard, mainly including two core indicators: line of sight range and gaze duration. Among them, the line of sight range is used to characterize the angular difference between the line of sight of the passer-by and the center of the advertising signboard, reflecting whether the passer-by pays attention to the advertising content; the gaze duration indicates the length of time the passer-by continues to pay attention to the advertising content, reflecting the degree of interest of the passer-by in the advertising content. In the embodiment of the present application, the line of sight range can be understood as an angle value. When the angle value is less than a preset threshold, it indicates that the line of sight of the passer-by falls within the effective viewing range of the advertisement; the gaze duration can be understood as a time accumulation value, and the system will record the continuous stay time of the passer-by within the line of sight range. The acquisition of visual features is used to identify potential target objects interested in the advertising content, and provide a behavioral basis for the subsequent optimization of the advertising content.

[0027] Correspondingly, group characteristics refer to the attribute characteristics of the passerby group in terms of demographics and spatial distribution, mainly including the number of passers-by, age group, gender ratio and group density. Among them, the number of passers-by reflects the size of the flow of people in the current area; the age group is used to distinguish audience groups of different age groups; the gender ratio indicates the proportion of males and females in the current group; and the group density is used to characterize the ratio of the number of target objects to the total number of passers-by. In the embodiment of the present application, the above-mentioned group characteristics can be understood as a set of statistical indicators, and the system describes the overall characteristics of the current audience group by calculating the values ​​of these indicators in real time. The acquisition of group characteristics is used to analyze the audience's population portrait, guide the system to select the advertising content that best suits the current group attributes, and realize the precision and personalization of advertising delivery.

[0028] Based on the above embodiment, as an optional embodiment, in step 102: the step of determining one or more target advertisements based on the visual features and the group features may further include the following steps: Step 201: Determine a target object that is interested in the advertising sign based on visual features.

[0029] The target object refers to individual passers-by who show continuous attention and high interest in the advertising content within the area of ​​the advertising sign. In the embodiment of the present application, the target object can be understood as a specific passer-by who meets both the sight range matching and gaze duration requirements. Determining the target object is mainly used to provide an effective audience basis for the intelligent adjustment of advertising content. By continuously tracking and analyzing the behavioral characteristics of the target object, the system can accurately evaluate the attractiveness of the currently playing content and provide data support for the subsequent optimization of advertising delivery strategies.

[0030] Based on the above embodiment, as an optional embodiment, the visual features include the sight range and the gaze duration, and the sight range is used to characterize the angle difference between the sight direction of the passerby and the center of the advertising sign. For step 201: based on the visual features, the step of determining the target object interested in the advertising sign can also specifically include the following steps: Step 301: Based on the sight range, determine the matching degree between the advertising sign and the sight range of the passerby.

[0031] Among them, the line of sight range refers to the spatial relationship parameter between the direction of the pedestrian's line of sight and the center position of the advertising sign, which mainly reflects the visual focus state of the pedestrian when watching the advertising content. In the embodiment of the present application, the line of sight range can be understood as the angle θ formed between the pedestrian's line of sight direction vector and the center line vector of the advertising sign. The smaller the angle, the closer the pedestrian's line of sight is to the center position of the advertising content. For example, when the pedestrian looks directly at the advertising sign, the angle θ is close to 0 degrees; when the pedestrian squints at the advertising sign at a 45-degree angle, the line of sight range obviously deviates from the optimal viewing angle. The line of sight range is used to evaluate the quality of visual interaction between pedestrians and advertising signs, and is a basic parameter for judging whether the pedestrian is actually paying attention to the advertising content.

[0032] Exemplarily, the sight range matching degree can be calculated by a first preset formula, which is expressed as: ; In the formula, Indicates the degree of sight range matching, Indicates the angle difference between the direction of the passerby's sight and the center of the advertising sign. It indicates the maximum allowable sight deviation angle, that is, the maximum allowable angle between the passerby's sight range and the center of the advertising sign.

[0033] in, ; In the formula, Indicates obtaining the sight vector of passers-by. Indicates getting the normal vector of the ad logo.

[0034] Specifically, the system first needs to obtain the passerby's sight vector and the normal vector of the advertising sign. The facial images of passersby are collected by high-definition cameras deployed around the advertising sign, and the eye position and pupil orientation of the passerby are extracted using computer vision algorithms to determine the spatial direction of the sight vector. At the same time, based on the installation position and orientation information of the advertising sign, the system can obtain the normal vector perpendicular to the advertising display plane. After determining these two vectors, the angle between the passerby's sight direction and the center of the advertising sign can be calculated through the vector dot product operation and the ratio of the vector modulus length.

[0035] After obtaining the sight angle, the system introduces a preset maximum sight deviation angle as a standardized reference value, which is usually set to 30 degrees, which is determined based on the physiological characteristics of the comfortable viewing range of the human eye. The sight range matching degree of a single passerby is calculated by the first preset formula. When the passerby looks directly at the advertising content, the matching degree reaches the maximum value; as the sight deviation angle increases, the matching degree decreases linearly; when the deviation angle exceeds the maximum allowable value, the matching degree drops to zero. The above calculation method not only conforms to the gradual characteristics of visual perception, but also can effectively filter out invalid samples with obvious sight deviations.

[0036] For example, for multiple passers-by within the range of the advertising sign, the sight range matching degree can be expressed as: ; Where N is the number of passers-by within the range of the advertising sign. represents the sight angle of the i-th passerby.

[0037] Specifically, considering that in actual scenes there are usually multiple passers-by watching advertisements at the same time, the system has designed a method to calculate the group's sight range matching degree. For multiple passers-by within the range of the advertising sign, the system calculates the arithmetic mean of all individual matching degrees to obtain the overall sight range matching degree. This calculation method not only retains the influence of individual sight characteristics, but also reflects the overall viewing status of the group, providing more comprehensive data support for evaluating the effectiveness of advertising.

[0038] Step 302: Based on the gaze duration, determine the matching degree between the advertisement currently broadcasted by the billboard and the gaze duration of the passerby.

[0039] The gaze duration refers to the duration of time that a passerby keeps his / her sight within the effective viewing range of the advertising sign within a specific time period. In the embodiment of the present application, the gaze duration can be understood as the cumulative time that the system detects that the match degree of the passerby's sight range continues to maintain a high level.

[0040] Specifically, when the passerby's line of sight continues to meet the matching conditions, the system will record and accumulate their effective viewing time; once the line of sight is detected to be significantly deviated or interrupted, the time accumulation will stop and the current value will be recorded. The gaze duration is used to quantify the degree of continuous attention of passersby to the advertising content, and is an important indicator for evaluating the attractiveness of advertising and the effectiveness of content.

[0041] Step 303: Based on the sight range matching degree and the gaze duration matching degree, determine the passerby's interest level value for the advertising signboard, and determine the passerby whose interest level value is greater than a threshold as a target object.

[0042] Specifically, in the advertising delivery system, a single visual feature is difficult to fully reflect the real interest of passers-by in the advertising content. In order to accurately identify the target object, the embodiment of the present application proposes a comprehensive evaluation method based on the matching degree of sight range and the matching degree of gaze duration, and realizes the accurate quantification of the interest degree of passers-by through multi-dimensional feature fusion.

[0043] Specifically, the system first needs to obtain two basic indicators: the match of the passerby's sight range and the match of the gaze duration. The sight range match reflects the spatial correspondence between the passerby's sight and the advertising content, while the gaze duration match represents the passerby's continuous attention to the advertising content. By setting different weight coefficients, the system performs a weighted fusion calculation on these two matching indicators to obtain the passerby's interest in the advertising sign.

[0044] After obtaining the interest level value, the system compares it with the preset interest level threshold. When the passerby's interest level value exceeds the threshold, the system marks it as a target object, indicating that the passerby has shown a clear willingness to pay attention to the advertising content.

[0045] Exemplarily, the interest level value can be calculated by a second preset formula, expressed as: ; In the formula, Indicates the interest level value, Indicates the degree of sight range matching, represents the matching degree of fixation duration, Indicates the duration of gaze, Represents the maximum fixation duration, used for normalization.

[0046] In different application scenarios, the importance of sight range and gaze duration may vary, and the weight coefficients α and β need to be adjusted to meet specific needs. The present application embodiment provides two typical application scenarios and their corresponding weight configuration schemes.

[0047] In a fast response scenario, such as the real-time push of promotional information in a shopping mall, the system pays more attention to capturing the immediate attention of passers-by. At this time, the advertising content is usually short text and eye-catching price information, requiring the system to quickly identify potential customers who are interested in promotional information. In this scenario, the system sets the weight coefficient α of the line of sight to 0.6 and the weight coefficient β of the gaze duration to 0.4. The above configuration emphasizes the immediate response characteristics of the line of sight matching. As long as the passer-by's line of sight turns to the advertising content, the system can capture this attention behavior in time, thereby achieving accurate push of promotional information.

[0048] In deep content scenarios, such as brand image promotion, the system pays more attention to the degree of sustained attention of passers-by. At this time, the advertising content often contains richer brand stories or product details, and it takes a long time to fully convey the information. In this scenario, the system sets the weight coefficient α of the line of sight to 0.3 and the weight coefficient β of the gaze duration to 0.7. This configuration emphasizes the importance of matching the gaze duration, requiring passers-by not only to focus their eyes on the advertising content, but also to maintain a long enough viewing time to be identified as a target object of real interest.

[0049] Step 202: Determine one or more target advertisements based on the group characteristics of the target objects.

[0050] Specifically, in the advertising delivery system, merely identifying the target object is not enough to achieve accurate advertising delivery. It is also necessary to deeply analyze the group characteristics of the target object to select the advertising content that best suits the current audience group. The embodiment of the present application proposes an advertising content intelligent matching method based on group characteristics, which realizes dynamic optimization of advertising content by analyzing the number, age group, gender ratio, and group density of the target object.

[0051] Based on the above embodiment, as an optional embodiment, the group characteristics include the number of passers-by, age group, gender ratio and group density. In step 202, the step of determining one or more target advertisements based on the group characteristics of the target object may further include the following steps: The number of pedestrians refers to the total number of all pedestrians that appear simultaneously within the area of ​​the advertising sign. In the embodiment of the present application, the number of pedestrians can be understood as the number of valid human targets identified by the system through the target detection algorithm within a preset time window, and this value will be dynamically updated over time.

[0052] Age group refers to the age distribution interval characteristics of passers-by. In the embodiment of the present application, the age group can be understood as the age estimation result obtained by the system after analyzing the facial features of passers-by, which is usually divided into three main intervals: youth (18-35 years old), middle-aged (36-55 years old) and elderly (over 56 years old). For example, the system may detect that the current target object is distributed in a situation where young people account for 60%, middle-aged people account for 30%, and elderly people account for 10%. The age group is used to guide the system to select advertising content suitable for specific age groups.

[0053] The gender ratio refers to the distribution of the number of males and females in a group of passersby. In the embodiment of the present application, the gender ratio can be understood as the ratio of the number of males and females calculated by the system after identifying the gender characteristics of passersby. The gender ratio is used to help the system select advertising content that is more suitable for the current gender composition.

[0054] Step 301: Obtain the group density of target objects. The group density is used to represent the ratio of the number of target objects to the number of passers-by.

[0055] Among them, group density refers to the ratio of the number of target objects to the total number of passers-by. In the embodiment of the present application, group density can be understood as the ratio of the number of target objects identified by the system that show continuous attention to the advertising content to the current total number of passers-by within a specific time window. Group density is used to evaluate the attractiveness level of the current advertising content and is an important basis for the system to determine whether it is necessary to adjust the advertising delivery strategy.

[0056] Exemplarily, the population density can be calculated by a third preset formula, which is expressed as follows: ; In the formula, represents the group density of the target object in front of the billboard at time t, Indicates the number of target objects, Indicates the number of passers-by.

[0057] Step 302: If the group density is greater than a threshold, one or more target advertisements are determined based on the age group and gender ratio.

[0058] Specifically, when the advertising content obtains a high group density, it means that the current content has successfully attracted the attention of passers-by. In order to maximize the advertising effect, the most suitable subsequent advertising content can be selected by analyzing the crowd portrait characteristics of the target object.

[0059] Specifically, the system first determines whether the current group density exceeds the preset threshold. When the group density is greater than the threshold, the system immediately begins to analyze the target object's age distribution and gender ratio characteristics. Through deep learning algorithms, the system can extract age and gender information from the target object's facial features and generate group portrait data.

[0060] After obtaining these group characteristic data, the system will perform matching screening in the advertising content library. Each advertisement in the advertising content library is pre-labeled with applicable target population characteristics, including age group preferences and gender tendencies. The system uses a weighted scoring method to calculate a matching score for each candidate advertisement based on the group characteristics of the current target object. For example, when it is detected that the target object is mainly young women, the system will give priority to fashion and beauty advertising content; when the target object presents the characteristics of a middle-aged group and a balanced ratio of men and women, the system may tend to choose home and automotive advertising content.

[0061] To ensure the continuity of advertising delivery, the system will also consider the relevance between advertising content. On the basis of ensuring the matching of group characteristics, it will give priority to ads with similar themes or content formats to the current ad, avoiding overly abrupt content switching. The system will eventually select one or more of the highest-scoring candidate ads as target ads, and deliver them in sequence according to the preset playback strategy.

[0062] Step 302: If the group density is less than or equal to the threshold, determine the group density change rate within a preset time period, and determine one or more target advertisements based on the group density change rate.

[0063] The population density change rate refers to the changing trend of the proportion of target objects within the advertising sign area within a preset time period. The population density change rate is used to evaluate the attractiveness trend of advertising content and guide the real-time adjustment of the delivery strategy. A positive change rate indicates that the advertising content is gradually attracting more attention, and the system may choose to maintain the current delivery strategy; a negative change rate prompts the system to adjust the advertising content in a timely manner to avoid continuous loss of viewers.

[0064] Exemplarily, the population density change rate can be calculated by a fourth preset formula, which is expressed as follows: ; In practical applications, the density change rate can be approximated using discrete time intervals: ; In the formula, Indicates a time interval.

[0065] when When , it indicates that the group density is increasing, that is, the number of target objects is increasing relative to the total number of passers-by; when , it indicates that the group density is decreasing, that is, the number of target objects is decreasing relative to the total number of passers-by.

[0066] Specifically, the system uses an approximate calculation method of discrete time intervals to obtain the rate of change of group density. In actual applications, the system will set a fixed sampling time interval and record the current group density D(t) at each sampling moment. By calculating the difference in group density between two adjacent sampling moments and dividing it by the sampling time interval, the rate of change of group density ΔD(t) at the current moment can be obtained. For example, if the group density at moment t is 0.25, and the group density 10 seconds ago (t-Δt moment) is 0.2, then the current rate of change of group density is about 0.005 per second, indicating that the advertising content is steadily attracting more attention.

[0067] The system evaluates the changing trend of the attractiveness of the advertising content by judging the positive or negative change rate of the group density. When the calculated change rate is greater than 0, it indicates that the number of target objects is increasing relative to the total number of passers-by, indicating that the current advertising content is constantly attracting new audiences; when the change rate is less than or equal to 0, it indicates that the proportion of target objects is decreasing or remaining stable, and it may be necessary to consider adjusting the advertising content. In order to avoid interference caused by instantaneous fluctuations, the system usually combines the data of multiple sampling points for smoothing to obtain a more stable judgment of the change trend.

[0068] When the population density is less than or equal to the threshold, the effect of advertising can be measured by an objective function that depends on the population density change rate. and the type of ads played , assuming there are several ads Choose from, each ad has its popularity , and the matching function suitable for different population density change rates .

[0069] Then the objective function can be expressed as: ; Correspondingly, in When the density is high, it is suitable to play long, static advertisements to attract fewer but potentially interested passers-by. The matching function can be reflected as the absolute value of the density change rate, which can be expressed as: ; In the formula, and It's an advertisement A specific parameter that indicates the ad's sensitivity to density changes.

[0070] exist When you are watching a video, you can choose a medium-length ad that is suitable for the current audience, such as a product introduction or brand promotion. The matching degree can be set to a constant: ; Specifically, the system first creates a feature attribute profile for each candidate ad, including its popularity and sensitivity parameters to changes in crowd density. Popularity reflects the basic appeal of the ad content, which can be determined through historical playback data and user feedback; sensitivity parameters characterize the adaptability of the ad content to different viewing environments and need to be set according to the ad type and presentation format. For example, a well-crafted brand story ad may have a high popularity, while its sensitivity parameters reflect the performance characteristics of the ad under different crowd densities.

[0071] The system uses an objective function to evaluate the real-time playback effect of each advertisement. When a negative value is detected in the rate of change of group density, it indicates that the current viewing population is decreasing, and the system tends to select long-duration, static-type advertising content. This type of advertisement is evaluated for effectiveness through a specific matching function, where a higher basic parameter value indicates that the advertisement has a stronger basic appeal, while the sensitivity parameter controls the degree of responsiveness of the advertisement to changes in group density. When the rate of change of group density is close to zero, the system will select medium-duration advertising content, such as product introductions or brand promotional videos. At this time, the matching degree is evaluated using a fixed value to ensure that the most suitable advertising content for the current stable viewing environment is selected.

[0072] In order to select the best advertisement, the main goal is to maximize the objective function, so the optimization problem can be defined as: ; In order to solve the problem of dependency between advertisements, especially in the scenario of advertisement sequence optimization, we can construct a state transition equation, consider the changes in population density at different time points in the future, and optimize the cumulative effect of advertisements played over a period of time. In T time periods, the population density change rate changes with time. , then the state transfer equation is: ; In the formula, Indicates the selection of advertisement at time t The optimal cumulative objective function value after t−1 is calculated, j represents the advertisement played at the previous time point t−1, and finally the optimal advertisement sequence selected in each time period t is backtracked.

[0073] Specifically, the matching degree of the current group density change is used as an evaluation indicator. This optimization model not only considers the immediate effect of a single advertisement, but also the coherence between advertisement sequences. For example, after playing a brand story advertisement, playing a related product introduction immediately afterwards may achieve better communication effects.

[0074] In order to solve the dependency problem between advertisements, the system uses a dynamic programming algorithm to construct a state transition equation. In each time period, the system needs to select the optimal advertisement content based on the previous moment's playback status and the current changes in group density. By recording the cumulative effect of selecting different advertisements at each moment, the system can find the optimal advertisement playback sequence within the entire planning time period. For example, if it is detected that the previous advertisement attracted more young viewers, the system will give priority to relevant advertisement content suitable for young groups in subsequent time periods.

[0075] The system determines the final ad play sequence through a backtracking algorithm. Starting from the last time point, the system gradually deduces the ad content that should be played in each time period based on the optimal choice recorded by the state transition equation. This method ensures that not only the choice at each time point is locally optimal, but also the cumulative effect of the entire ad sequence reaches the global optimal.

[0076] Step 103: Obtain the movement trajectory of the passerby in the target image.

[0077] The path trajectory refers to the sequence of spatial positions of passers-by in the target area that changes over time. In the embodiment of the present application, the path trajectory can be understood as the movement trajectory of the passers-by represented by a two-dimensional coordinate system, including the position coordinates, movement speed and direction of the passers-by at a specific time point. The path trajectory is used to characterize the relative position relationship between the displacement of the passers-by and each advertising sign.

[0078] Specifically, the system first needs to calculate the real-time distance between the passerby and the advertising sign. Each advertising sign has a fixed position coordinate in a two-dimensional plane, while the position coordinates of the passerby change over time. By calculating the Euclidean distance between the current position of the passerby and the position of the advertising sign, the system can obtain the spatial distance between the two in real time. This distance calculation method is intuitive and efficient, and provides reliable data support for the system to monitor the positional relationship between passers-by and advertising signs.

[0079] Based on the location data, the system further analyzes the movement characteristics of passers-by. The movement trajectory of passers-by is essentially the trajectory curve of their position changing over time. By performing time differentiation on the location data, the instantaneous speed of the passers-by can be obtained. The system continuously collects the location data of passers-by through the deployed camera network, and obtains complete movement trajectory information after processing.

[0080] Step 104: Control one or more advertising signs to play target advertisements based on the travel trajectory.

[0081] After determining the target advertisement, the system needs to control the advertisement playback strategy according to the characteristics of the pedestrians' movement trajectory to achieve the best advertisement display effect. The positional relationship between pedestrians and different advertisement signs changes constantly during their movement. The above dynamic changes directly affect the viewing effect of the advertisement content. Therefore, the playback order and duration of advertisements can be optimized by analyzing the movement characteristics of pedestrians.

[0082] Based on the above embodiment, as an optional embodiment, in step 104: the step of controlling one or more advertising signs to play target advertisements based on the travel trajectory may further include the following steps: Step 401: Obtain the relative distance between each passerby and each advertising sign based on the movement trajectory.

[0083] For example, assuming that the area where the advertising sign is located is a two-dimensional plane, the trajectory of the passerby can be represented by the change of its displacement over time: ; In the formula, The position of the jth advertising sign is expressed as a two-dimensional coordinate ( ), represents the pedestrians approaching the advertising sign at time t The distance of the pedestrian at time t is expressed as a two-dimensional coordinate ( ).

[0084] The trajectory of a passerby can be expressed as its position The trajectory changes over time t. The position change data of the passerby can be obtained, and the speed of the trajectory change It is expressed as: .

[0085] Step 402: Determine a distance weight based on the relative distances of the advertising signs. The distance weight is used to characterize the possibility of a passerby seeing the corresponding advertising sign based on the corresponding movement trajectory.

[0086] Among them, the distance weight refers to a quantitative indicator of the possibility of passers-by viewing the content of the advertising sign, which comprehensively considers the spatial distance between the passers-by and the advertising sign and the characteristics of the direction of travel. In the embodiment of the present application, the distance weight can be understood as the product of the basic spatial weight and the direction adjustment factor, where the basic spatial weight is in the form of the inverse of the distance to reflect the impact of the distance on the viewing possibility, and the direction adjustment factor characterizes the degree of match between the passer-by's line of sight and the advertising content through the cosine value of the angle between the passer-by's direction of travel and the direction of the advertising sign.

[0087] For example, for each advertising sign Calculate a weight related to the distance of the passerby , the weight can be calculated based on the inverse of the distance. The closer the distance, the greater the weight, expressed as: ; In the formula, Used to prevent the denominator from being 0.

[0088] Furthermore, the angle of travel of passers-by can be The play order can be adjusted by the angle between the pedestrian and the advertising sign. The direction of travel of the pedestrian can be determined by the change in the current position and the position at the next moment, expressed as: ; The angle between the advertising sign and the current position of the passerby is expressed as: ; The angle between the pedestrian's direction of travel and the advertising sign is expressed as: ; Specifically, the above calculation method ensures that the closer the advertising sign is, the higher the weight value is. In order to avoid calculation anomalies when passers-by are too close to the advertising sign, the system introduces a small smoothing factor in the denominator. For example, when a passer-by is 2 meters away from an advertising sign, its basic weight value may be 0.4; when the distance increases to 10 meters, the weight value may decrease to 0.09. The weight calculation method based on the inverse of the distance intuitively reflects the influence of spatial distance on the possibility of advertising viewing.

[0089] After obtaining the basic distance weight, the system further considers the characteristics of the pedestrian's movement direction. By analyzing the changes in the pedestrian's position at adjacent moments, the system can calculate the pedestrian's real-time direction of travel. Specifically, the system uses the pedestrian's position coordinates at the current moment and the next moment to calculate the travel angle through the inverse tangent function. At the same time, the system also calculates the azimuth of the advertising sign relative to the pedestrian's current position, which reflects the position of the advertising sign in the pedestrian's field of vision. By calculating the difference between these two angles, the system obtains the angle between the pedestrian's direction of travel and the direction of the advertising sign.

[0090] The calculation of the angle enables the system to more accurately assess the likelihood of passersby viewing the advertising content. For example, when a passerby moves directly toward the advertising sign, the angle is close to 0 degrees, indicating that the passerby is more likely to notice the advertising content; when a passerby approaches the advertising sign at a 45-degree angle, the likelihood of viewing decreases accordingly even if the distance is the same; if a passerby moves with his back to the advertising sign, that is, the angle is close to 180 degrees, the likelihood of viewing is extremely low even if the distance is closer.

[0091] Furthermore, the smaller the angle, the more likely passersby are to see the advertising sign. Therefore, the weight can be adjusted to: ; Step 403: Determine the play order of one or more advertising signboards based on the distance weights, and play one or more target advertisements on each advertising signboard based on the play order.

[0092] Specifically, the order of playing the advertising signs can be sorted based on weights. Advertising signs with higher weights are played earlier. This can be expressed as: ; Specifically, after obtaining the adjusted weight values ​​of each advertising sign, the system needs to set the order of advertisement playback to achieve the optimal configuration of advertising resources. Since the attention and viewing time of passers-by are limited, a reasonable playback order is crucial to improving the effect of advertising. Therefore, the embodiment of the present application proposes a method for determining the playback order based on weight maximization, which improves the dissemination effect of advertising content by dynamically optimizing the playback order.

[0093] Specifically, the system first prioritizes all advertising signs based on the adjusted distance weights. By maximizing the weight principle, the system selects the advertising sign with the highest weight value as the primary playback position. This selection mechanism ensures that advertising content is played first on the advertising sign that passers-by are most likely to notice. For example, when a passer-by is detected approaching an advertising sign and the direction of travel is basically the same as that of the sign, the sign may obtain the highest adjusted weight value, and the system will give priority to playing advertisements on this sign.

[0094] After determining the playback order, the system starts to play the target advertisement on each billboard in order of priority. This playback process is dynamic. The system will continuously monitor the position changes and movement directions of passers-by, and update the weight value and playback order of each billboard in real time. For example, when the viewing conditions of a billboard improve due to the movement of passers-by, the system may increase the playback priority of the billboard; conversely, if the viewing conditions deteriorate, it may reduce its priority or suspend playback.

[0095] When there are multiple target ads to be played, the system will intelligently match them based on the characteristics of the ad content and the viewing behavior of passers-by. For ad signs with higher weights, the system tends to play more important or more suitable ad content for immediate viewing; for signs with lower weights, it may choose to play secondary information or cyclical content. The above differentiated content distribution strategy ensures that important advertising information can be delivered to the target audience at the best time and location.

[0096] Based on the above embodiment, as an optional embodiment, after determining the playback order of the advertising signboards, the following steps may also be included: Step 501: Determine the dwell time of the passerby at the next advertising sign based on the relative distance of the passerby to each advertising sign.

[0097] For example, the residence time of a passerby near the advertising signboard can be calculated by the distance between the passerby and the advertising signboard being less than a threshold value, which can be expressed as: ; In the formula, {1} indicates the function, which takes 1 when the condition is met, otherwise it takes 0; and It is the time frame that passers-by are near the advertising sign.

[0098] Specifically, the system first defines a distance threshold to determine whether a passerby is within the effective viewing range of the billboard. The setting of this threshold takes into account factors such as the billboard size, installation height, and clarity of the displayed content. For example, for a standard-sized LED advertising screen, the system may set the distance threshold to 5 meters, within which passersby can clearly see the advertising content and have effective visual interaction.

[0099] After determining the distance threshold, the system starts to monitor the relative distance between the passerby and the advertising sign in real time. When the passerby is detected to enter the effective viewing range, the system starts the time accumulation count; when the passerby leaves the range, the system stops counting and records the accumulated time. This accumulation process takes into account the situation that the passerby may enter and leave the effective range multiple times. Through continuous distance monitoring and time accumulation, the system can obtain the total residence time of the passerby in front of a specific advertising sign.

[0100] The system uses an indicator function to mark whether a passerby is within the effective viewing range. When the distance between the passerby and the advertising sign is less than the preset threshold, the indicator function takes a value of 1, indicating that the passerby is within the effective range; when the distance is greater than the threshold, the indicator function takes a value of 0, indicating that the passerby has left the effective range. By integrating the indicator function within a specific time interval, the system can accurately calculate the actual residence time of the passerby.

[0101] Step 502: Adjust the playing duration of one or more target advertisements based on the dwell time.

[0102] For example, the advertisement playing time may be associated with the dwelling time of passers-by, so a minimum playing time may be set and extended according to the dwelling time. The process may be expressed as: ; Where k is a proportional coefficient used to adjust the scalability of the playback duration.

[0103] Specifically, the system can set a minimum playback time, which ensures that the advertising content can fully convey the core information. The setting of the minimum playback time takes into account the complexity and amount of information in the advertising content. For example, a simple promotional message may be set to 10 seconds, while a brand image advertisement may require a benchmark time of more than 20 seconds. This benchmark setting provides a basic guarantee for advertising playback, ensuring that the necessary advertising information can be delivered even when passers-by pass by quickly.

[0104] Based on the benchmark duration, the system introduces a proportional coefficient to achieve dynamic extension of the playback duration. This proportional coefficient determines the system's responsiveness to the pedestrian's lingering behavior. A larger proportional coefficient means that the playback duration is more sensitive to the lingering time. For example, when the proportional coefficient is set to 0.5, if it is detected that the pedestrian stays in front of the advertising sign for an additional 20 seconds, the system will extend the playback duration by 10 seconds accordingly. This linear extension mechanism maintains the simplicity of playback control while achieving dynamic adaptation to viewing behavior.

[0105] The system will also dynamically adjust the scale factor according to different scenarios and ad types. In high-mobility places such as shopping malls, the system may use a smaller scale factor to prevent a single passerby from watching for too long and affecting other potential viewers; in low-mobility places such as airport waiting areas, a larger scale factor can be used to provide more complete ad content to interested passersby. This scenario-based parameter adjustment ensures the flexibility and adaptability of playback control.

[0106] Please refer to Figure 2 , Figure 2 Another advertising signboard playback method based on passerby vision provided in an embodiment of the present application is applied to an advertising signboard, and the advertising signboard is communicatively connected with one or more other advertising signboards in a target area. Specifically, the method may further include the following steps: Step 601: Acquire a target image of passers-by within a region; Step 602: Obtain visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and group features; Step 603: Acquire the track of the passerby in the target image; Step 604: Control one or more other advertising signs to play target advertisements based on the travel trajectory.

[0107] It should be noted that the two methods in the embodiments of the present application have different execution entities. Figure 1 The execution subject of the embodiment shown is an edge server, which uniformly manages and controls multiple advertising signs in the target area; Figure 2 The execution subject of the illustrated embodiment is an advertising signboard, with a single advertising signboard serving as a master control device to coordinate and control the playback behaviors of other advertising signboards.

[0108] Specifically, when the edge server is used as the execution subject, it directly obtains the information of passers-by in the coverage area of ​​all advertising signs in the target area, realizing global monitoring and unified management of the entire target area. The edge server can process data from multiple advertising signs at the same time and make playback decisions based on global information, which is conducive to achieving a wider range of advertising coordination.

[0109] When the advertising sign acts as the execution subject, it mainly analyzes and makes decisions based on the information of passers-by in its own area, and coordinates the playback behavior through communication connections with other advertising signs. This distributed control method reduces the dependence on the central server and improves the flexibility and scalability of the system. Even if an advertising sign fails, it will not affect the operation of the entire system.

[0110] The above two execution entities have their own advantages, and the appropriate implementation method can be selected according to the needs of the actual application scenario. For example, in places with relatively concentrated spaces such as shopping malls, edge servers can be used as the execution entity; while in decentralized scenarios such as blocks, advertising signs can be used as the execution entity.

[0111] It should be noted that the specific implementation process of the above method can refer to the relevant description of steps 101 to 104 in the above embodiment, which will not be repeated here.

[0112] Please refer to Figure 3 , Figure 3 A schematic diagram of the structure of an edge server provided in an embodiment of the present application, the edge server comprising: A first target image acquisition module is used to acquire a target image of a passerby within the advertising sign area; A first target advertisement determination module, configured to obtain visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; A first movement track acquisition module, used to acquire the movement track of the passerby in the target image; The first target advertisement playing module is used to control one or more of the advertisement signs to play the target advertisements based on the travel trajectory.

[0113] Based on the above embodiment, as an optional embodiment, the first target advertisement determination module is further used to determine the target object interested in the advertising sign based on the visual features; and determine one or more target advertisements based on the group characteristics of the target object.

[0114] On the basis of the above embodiment, as an optional embodiment, the first target advertisement determination module is further used to determine the matching degree between the advertising signboard and the visual range of the passerby based on the visual range; determine the matching degree between the advertisement currently broadcast by the advertising signboard and the visual range of the passerby based on the gaze duration; determine the interest level of the passerby in the advertising signboard based on the visual range matching degree and the gaze duration matching degree, and determine the passerby whose interest level value is greater than a threshold as the target object.

[0115] Based on the above embodiment, as an optional embodiment, the first target advertisement determination module is further used to obtain the group density of the target object, and the group density is used to characterize the ratio of the number of the target objects to the number of passers-by; if the group density is greater than a threshold, one or more target advertisements are determined based on the age group and the gender ratio; if the group density is less than or equal to a threshold, the rate of change of the group density within a preset time period is determined, and one or more target advertisements are determined based on the rate of change of the group density.

[0116] On the basis of the above embodiment, as an optional embodiment, the first movement trajectory acquisition module is also used to obtain the relative distance between each of the passers-by and each of the advertising signs based on the movement trajectory; determine the distance weight based on the relative distance of each of the advertising signs, and the distance weight is used to characterize the possibility of the passer-by seeing the corresponding advertising sign based on the corresponding movement of the movement trajectory; determine the playback order of one or more of the advertising signs based on the distance weight, and play the one or more target advertisements on each of the advertising signs based on the playback order.

[0117] Based on the above embodiment, as an optional embodiment, the first movement trajectory acquisition module is also used to determine the residence time of the passerby at the next advertising sign based on the relative distance of the passerby to each of the advertising signs; and adjust the playback duration of one or more of the target advertisements based on the residence time.

[0118] Please refer to Figure 4 , Figure 4 A schematic diagram of the structure of an advertising sign provided in an embodiment of the present application, the advertising sign comprising: The second target image acquisition module is used to acquire target images of passers-by within the area; A second target advertisement determination module, configured to obtain visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; A second movement track acquisition module, used to acquire the movement track of the passerby in the target image; The second target advertisement playing module is used to control one or more other advertisement signboards to play each of the target advertisements based on the movement trajectory.

[0119] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0120] The embodiment of the present application also provides a computer storage medium, which can store multiple instructions. The instructions are suitable for being loaded by a processor and executed by the advertising sign playing method based on passerby vision such as the above-mentioned embodiment. The specific execution process can refer to the specific description of the embodiment shown, which will not be repeated here.

[0121] The present application also discloses an electronic device. Figure 5 , Figure 5 The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

[0122] The communication bus 502 is used to realize the connection and communication between these components.

[0123] The user interface 503 may include a display interface and a camera interface. Optionally, the user interface 503 may also include a standard wired interface and a wireless interface.

[0124] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0125] Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Optionally, the processor 501 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 501 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface diagrams and applications, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501, and it can be implemented separately through a chip.

[0126] Among them, the memory 505 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 may also be optionally at least one storage device located away from the aforementioned processor 501. Refer to Figure 5 , the memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of an advertising sign playing method based on passerby vision.

[0127] exist Figure 5In the electronic device 500 shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call the application program stored in the memory 505 for a method of playing an advertising sign based on the vision of passers-by. When executed by one or more processors 501, the electronic device 500 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0128] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0130] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0131] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0132] If 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 memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0133] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0134] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for playing advertising signs based on passerby vision, characterized in that: Applied to an edge server, the edge server is in communication connection with one or more advertising signs in a target area, and the advertising sign playing method based on passerby vision includes: Acquire a target image of a passerby within the area of ​​the advertising signboard; Acquire visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; Obtaining the trajectory of the passerby in the target image; Based on the travel trajectory, one or more of the advertising signs are controlled to play the target advertisements.

2. The method for playing advertising signs based on passerby vision according to claim 1, characterized in that: The determining one or more target advertisements based on the visual features and the group features includes: Based on the visual features, determining a target object that is interested in the advertising sign; One or more targeted advertisements are determined based on the group characteristics of the target objects.

3. The method for playing advertising signs based on passerby vision according to claim 2, characterized in that: The visual features include a sight range and a gaze duration, wherein the sight range is used to characterize the angle difference between the sight direction of the passerby and the center of the advertising sign; and determining the target object interested in the advertising sign based on the visual features includes: Based on the sight range, determining the matching degree between the advertising sign and the sight range of the passerby; Based on the gaze duration, determining a matching degree between the advertisement currently broadcasted by the advertising sign and the gaze duration of the passerby; Based on the sight range matching degree and the gaze duration matching degree, the interest level value of the passerby in the advertising signboard is determined, and the passerby whose interest level value is greater than a threshold is determined as a target object.

4. The method for playing advertising signs based on passerby vision according to claim 2, characterized in that: The group characteristics include the number of passers-by, age group, gender ratio and group density; The step of determining one or more target advertisements based on the group characteristics of the target object includes: Acquire a group density of the target objects, where the group density is used to represent a ratio between the number of the target objects and the number of the passers-by; If the group density is greater than a threshold, determining one or more target advertisements based on the age group and the gender ratio; If the group density is less than or equal to a threshold, a change rate of the group density within a preset time period is determined, and one or more target advertisements are determined based on the change rate of the group density.

5. The method for playing advertising signs based on passerby vision according to any one of claims 1 to 4, characterized in that: The travel trajectory is used to characterize the relative position relationship between the displacement of the passerby and each of the advertising signs; The controlling one or more advertising signs to play the target advertisements based on the travel trajectory includes: Acquire the relative distance between each of the passers-by and each of the advertising signs based on the travel trajectory; Determining a distance weight based on the relative distance of each of the advertising signs, the distance weight being used to characterize the possibility of the passerby seeing the corresponding advertising sign based on the corresponding movement trajectory; The play order of one or more of the advertising signs is determined based on the distance weights, and the one or more target advertisements are played on each of the advertising signs based on the play order.

6. The method for playing advertising signs based on passerby vision according to claim 5, characterized in that: After determining the play order of one or more of the advertising signs based on the distance weight, the method further includes: Determining the dwelling time of the passerby at the next advertising sign based on the relative distance of the passerby to each of the advertising sign; The playing duration of one or more target advertisements is adjusted based on the dwell time.

7. A method for playing advertising signs based on passerby vision, characterized in that: Applied to an advertising signboard, the advertising signboard is in communication connection with one or more other advertising signboards in a target area, and the advertising signboard playing method based on passerby vision includes: Acquire target images of passers-by within the area; Acquire visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; Obtaining the trajectory of the passerby in the target image; Based on the travel trajectory, one or more other advertising signs are controlled to play each of the target advertisements.

8. An edge server, characterized in that: include: A first target image acquisition module is used to acquire a target image of a passerby within the advertising sign area; A first target advertisement determination module, configured to obtain visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; A first movement track acquisition module, used to acquire the movement track of the passerby in the target image; The first target advertisement playing module is used to control one or more of the advertisement signs to play the target advertisements based on the travel trajectory.

9. An advertising sign, characterized in that: include: The second target image acquisition module is used to acquire target images of passers-by within the area; A second target advertisement determination module, configured to obtain visual features and group features of passers-by in the target image, and determine one or more target advertisements based on the visual features and the group features; A second movement track acquisition module, used to acquire the movement track of the passerby in the target image; The second target advertisement playing module is used to control one or more other advertisement signboards to play each of the target advertisements based on the movement trajectory.

10. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

11. A computer storage medium, characterized in that: The computer storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images