CN113221002B

CN113221002B - Information display method and device

Info

Publication number: CN113221002B
Application number: CN202110546557.5A
Authority: CN
Inventors: 姚经纬; 杨文玉; 李书博
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2021-05-19
Filing date: 2021-05-19
Publication date: 2022-12-20
Anticipated expiration: 2041-05-19
Also published as: CN113221002A

Abstract

The specification discloses a method and a device for displaying information. An information display method is applied to display equipment and comprises the following steps: in the process of displaying first information, calling a camera module to acquire an environment image stream in the process of displaying the first information; determining a number of viewers of the first information based on the ambient image stream; carrying out crowd classification on the audiences to obtain crowd classification results of the audiences; determining the priority of each piece of information to be displayed based on the crowd classification result of each audience, and determining second information based on the priority; and displaying the second information after the first information is displayed.

Description

Information display method and device

Technical Field

The present disclosure relates to the field of internet technologies, and in particular, to a method and an apparatus for displaying information.

Background

With the development of technologies, scenes of information presentation are more and more increased in life, and for example, advertisements, news, recruitment information and the like can be presented for a user. How to reasonably display the information becomes the key point of business attention.

Disclosure of Invention

In view of this, the present specification provides a method and an apparatus for displaying information.

Specifically, the specification is realized through the following technical scheme:

a display method of information is applied to display equipment, and the method comprises the following steps:

in the process of displaying first information, calling a camera module to acquire an environment image stream in the process of displaying the first information;

determining a number of viewers of the first information based on the ambient image stream;

carrying out crowd classification on the audiences to obtain crowd classification results of the audiences;

determining the priority of each piece of information to be displayed based on the crowd classification result of each audience, and determining second information based on the priority;

and displaying the second information after the first information is displayed.

An information display device applied to display equipment, the device comprises:

the calling module is used for calling a camera module to collect an environment image stream in the first information display process;

an audience determination module to determine a number of audiences of the first information based on the ambient image stream;

the classification module is used for carrying out crowd classification on the audiences to obtain crowd classification results of the audiences;

the information determining module is used for determining the priority of each piece of information to be displayed based on the crowd classification result of each audience and determining second information based on the priority;

and the display module displays the second information after the first information is displayed.

An apparatus for displaying information, comprising:

a processor;

a memory for storing machine executable instructions;

wherein, by reading and executing machine-executable instructions stored by the memory that correspond to presentation logic of information, the processor is caused to:

and displaying the second information after the first information is displayed.

One embodiment of the present specification realizes that, in the process of displaying first information, a display device may invoke a camera module to collect an environment image stream corresponding to the first information, determine a plurality of audiences of the first information based on the environment image stream, classify the audiences, obtain crowd classification results of the audiences, determine the priority of each piece of information to be displayed based on the crowd classification results of the audiences, determine second information based on the priority, and display the second information after the first information is displayed. By adopting the method, the information can be reasonably displayed to the facing audience, the information publisher is helped to better popularize the information, and the benefit of information display is improved.

Drawings

FIG. 1 is a schematic diagram of an information presentation scenario shown in an exemplary embodiment of the present description;

FIG. 2 is a flow chart diagram illustrating a method for presenting information in accordance with an exemplary embodiment of the present disclosure;

FIG. 3 is a flow chart diagram illustrating another method of presenting information in accordance with an exemplary embodiment of the present disclosure;

fig. 4 is a hardware configuration diagram of a device in which an information presentation apparatus is located according to an exemplary embodiment of the present specification;

fig. 5 is a block diagram of an apparatus for presenting information, as shown in an exemplary embodiment of the present description.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with this description. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.

The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the description. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present description. The word "if" as used herein may be interpreted as "at" \8230; "or" when 8230; \8230; "or" in response to a determination ", depending on the context.

With the development of technology, scenes of information display are more and more common in life, and information display modes are more and more. In the case of advertisement, advertisement is a means of publicizing and widely delivering information to the public through a certain form of media for a specific need. The traditional advertisement display media are usually television, newspaper, and radio. With the development of computers, advertisements can be displayed through various intelligent devices, for example, the advertisements can be displayed to a user when the user watches videos by using a mobile phone, the advertisements can be displayed through advertisement display devices in a cell elevator room, the advertisements can be displayed at bus stops by using devices for displaying stop board information, and the like.

Although the methods of information presentation are varied, not all methods of information presentation are reasonably effective, and it is likely that some information is of no value and interest to the viewer. Therefore, how to reasonably display the information so as to reasonably display the information to the audience facing the information becomes the key point of business attention in helping the information publisher to better popularize the information.

An exemplary information presentation scenario is first described.

Referring to fig. 1, fig. 1 is a schematic diagram of an information presentation scenario according to an exemplary embodiment of the present disclosure.

The information display scenario in fig. 1 includes 3 exemplary information display devices and a server, where the server may be a server or a server cluster, and the information display device may be an information display device in an elevator hall, a bus stop information display device, and an outdoor large-screen information display device, and of course, the information display device shown in fig. 1 is only an exemplary description, and the method shown in this specification may also be applied to other devices besides the example shown in fig. 1.

The display equipment can be communicated with the server through a network, the server can issue a plurality of pieces of information to be displayed to each display equipment, the display equipment can display the information to be displayed after receiving the information, and the information to be displayed can be advertisements, news, broadcasts, notifications and the like or other information. The present embodiment does not specially limit the number of the display devices, the geographical locations, the types of the devices, and the manner of displaying the information.

Moreover, each display device may be further configured with a camera module, which may be a camera mounted on the device. In the process of displaying information, the display device may call a camera of the display device to acquire an environmental image stream, so as to determine a current audience based on the environmental image stream, and determine, according to the audience, a next piece of information to be displayed after the current information is displayed, which will be specifically described in detail in the following embodiments. Of course, if the display device itself does not have a camera, the camera in the surrounding environment of the display device may also be used to collect the environmental image stream, which is not particularly limited.

The following describes an embodiment of a method for displaying information provided in the present specification.

Referring to fig. 2, fig. 2 is a flowchart illustrating an information displaying method according to an exemplary embodiment of the present disclosure, where the information displaying method may be applied to an information displaying apparatus, such as the information displaying apparatus shown in fig. 1, and of course, may also be applied to other information displaying apparatuses other than the information displaying apparatus shown in fig. 1.

The information display method can comprise the following steps:

202, in the process of displaying first information, calling a camera module to acquire an environment image stream in the process of displaying the first information;

step 204, determining a plurality of audiences of the first information based on the environment image stream;

step 206, carrying out crowd classification on the audiences to obtain crowd classification results of the audiences;

step 208, determining the priority of each information to be displayed based on the crowd classification result of each audience, and determining second information based on the priority;

and step 210, displaying the second information after the first information is displayed.

The above steps are explained in detail below.

In this embodiment, information may be displayed by a plurality of display devices. The information may be advertisements, news, broadcasts, notifications, etc. Taking information as an example of an advertisement, the display device can be an outdoor advertisement screen, an elevator car advertisement display device, a bus stop advertisement display device and the like, and the type, the configuration place and the number of the display device are not particularly limited in the embodiment.

The display device may store a plurality of pieces of information to be displayed, and the information to be displayed may be issued by the server to the display device. And the plurality of information to be displayed can also have a default display order, so that the display device can display the information according to the default display order at the beginning. The first information and the second information described in the specification all belong to the plurality of information to be displayed. The terms "first" and "second" are used herein to distinguish the two pieces of information to be displayed, and should not be interpreted as having other technical limitations.

A camera module, such as a camera, can also be arranged on the display device. When the display device is displaying the first information, a camera can be called to collect an environment image stream corresponding to the first information, and then a plurality of audiences of the first information are determined based on the environment image stream.

For example, the presentation device may extract image frames from the ambient image stream and identify viewers from the image frames, and the specific method of identifying viewers will be described in detail in the following embodiments.

Of course, the display device may also send the acquired environmental image stream to the server, so that the server identifies a plurality of viewers from the environmental image stream. The display device may continuously send the acquired environment image stream to the server, and may also send a timestamp of the environment image stream and a playing time period of the first information, so that the server determines the environment image stream corresponding to the first information from the received environment image stream. Certainly, the display device may also invoke the camera to collect the environmental image stream and send the environmental image stream to the server after monitoring that the first information is played, which is not particularly limited.

In this embodiment, after determining the audience of the first information, the audience may be classified into groups to obtain a group classification result of each audience.

For example, for each viewer, an image frame in which the viewer is located may be extracted from the environmental image stream, and the viewer may be classified into groups based on the image frame. If the image frame may include many audiences, an image including only the audiences can be cut out from the image frame and then input into the crowd classification model to obtain the crowd classification result of the audiences. The crowd classification result can be multidimensional, and the crowd classification can be carried out according to three dimensions of gender, age and occupation. The crowd classification process may be performed by the display device or the server, which is not particularly limited.

In this embodiment, the priority of each piece of information to be displayed may be determined according to the crowd classification result of each viewer.

For example, the type of the crowd to which the information is directed may be preset for each information to be displayed. Then, matching the crowd classification result of the audience with the crowd type oriented by each piece of information to be displayed, and determining the priority of each piece of information to be displayed according to the matching degree of the audience and the crowd oriented information to be displayed, wherein the higher the matching degree is, the higher the priority is; the lower the degree of matching, the lower the priority.

Then, second information can be determined from all the information to be displayed based on the priority, and the second information is displayed after the first information is displayed.

As can be seen from the above description, in an embodiment of the present specification, in the process of displaying the first information, the display apparatus may invoke the camera module to collect an environment image stream corresponding to the first information, determine a plurality of audiences of the first information based on the environment image stream, classify the audiences, obtain a crowd classification result of each audience, determine a priority of each piece of information to be displayed based on the crowd classification result of each audience, determine the second information based on the priority, and display the second information after the first information is displayed.

By adopting the method, the display equipment can call the camera module to collect the environmental image stream in the process of displaying the first information, determine the audience watching the information at present based on the environmental image stream, and select the second information matched with the current audience from the information to be displayed according to the crowd type of the audience, so as to display the information to the audience facing the information reasonably, help the information issuing party to popularize the information better and improve the profit of information display.

An example of another method of displaying information provided in the present specification is described below.

Referring to fig. 3, fig. 3 is a schematic flowchart of another information displaying method according to an exemplary embodiment of the present disclosure, where the information displaying method may be applied to an information displaying apparatus.

The information display method can comprise the following steps:

step 302, a camera module is called to collect an environment image stream corresponding to the first information.

In this embodiment, the display device may be configured with a camera module, such as a camera. In the process of displaying the first information, the display device can acquire an environmental image stream around the display device through the camera module. The trigger time for calling the camera module to acquire the environmental image stream may be that the remaining playing time of the first information is less than a time threshold, and the time threshold may be 5 seconds or 3 seconds, and may be specifically preset.

In this example, since the viewers around the display device are mobile, the current ambient image stream can be acquired after the remaining playing time of the first information reaches the time threshold, and the viewers determined based on the ambient image stream will likely watch the second information after the playing of the first information is completed, thereby facilitating the display of the more matched second information to the viewers.

Step 304, determining a number of viewers of the first information based on the stream of ambient images.

In this embodiment, image frames may be extracted from the captured environmental image stream, and then the image frames may be identified to identify whether a pedestrian is present in the image frames. Whether a pedestrian is present or not can be recognized according to whether a human body shape, a human face, or the like is present in the image frame, for example.

Of course, in other embodiments, besides the pedestrian may be identified through the environmental image stream, the pedestrian may also be identified by combining other methods, for example, an infrared sensor, a voice sensor, and the like may also be used to assist in determining whether there is a pedestrian in the environment around the display device.

It should be noted that the pedestrian is not necessarily a walking (moving) person, but may be a stationary person, for example, a stationary person in an elevator car when the elevator car displays information.

In this embodiment, if there is a pedestrian, it may be further determined whether the pedestrian is a spectator. Taking the advertisement displayed by the elevator car as the target information as an example, the advertisement display device may shoot all pedestrians in the elevator car, but does not represent that all pedestrians are watching the advertisement, and it is meaningless to analyze pedestrians who do not watch the advertisement when analyzing the advertisement display effect, so that it is necessary to distinguish the audience who really watch the advertisement.

For example, the face orientation feature of the pedestrian may be further analyzed based on the image frames to determine the audience for the first information based on the face orientation feature. If the face of the pedestrian faces the camera, the pedestrian really watches the first information and is determined as the audience; if the face is oriented side-to-side and back-to-back with respect to the camera, it is likely that the pedestrian is not viewing the first information and is not determined to be a viewer. Of course, besides the face orientation feature of the pedestrian, the eye gazing direction of the pedestrian can also be analyzed, and if the camera is looked at in front, the pedestrian is determined to be the audience; otherwise, the pedestrian is determined not to be a spectator. The method of analyzing the face direction and the method of analyzing the eye gaze direction may refer to related art, and the present embodiment is not particularly limited thereto.

And step 306, carrying out crowd classification on the audiences to obtain crowd classification results of the audiences.

In this embodiment, after determining the audiences of the first information, the audiences can be classified into groups.

In one example, the wearing feature and appearance feature of the viewer can be analyzed from the image frames, so as to perform group classification. The demographic classification may be multidimensional, such as analyzing the audience's age, gender, occupation, and so on.

In another example, if the image frame may include many viewers, an image including only the viewers may be cut out from the image frame and then input into the crowd classification model to obtain the crowd classification result of the viewers.

The crowd classification model may be a neural network model or other models, and is not particularly limited. The method may specifically be obtained by training a large number of images as training samples and crowd classification results as sample labels, and reference may be made to related technologies.

Certainly, in other embodiments, other methods may also be used to classify the crowd, for example, the face of the audience may be recognized to identify the identity of the audience, and then the related information of the identity is obtained through the big data to classify the crowd, which is not illustrated here.

And 308, predicting the watching duration of each audience to the information to be displayed.

In this embodiment, the display device may predict, for each piece of information to be displayed, the viewing duration of each audience for the piece of information to be displayed, so as to determine, based on the predicted viewing duration, the piece of information to be displayed having a high audience interest level and display the piece of information to be displayed to the audience.

The watching time length of each audience to the information to be displayed can be predicted through the time length prediction model aiming at each information to be displayed.

In one example, for each viewer, the content features of the information to be presented and the crowd classification result of the viewer may be input into the first time length prediction model, and the viewing time length of the viewer may be predicted.

The content features of the information to be presented may represent the content included in the information to be presented, for example, the content features may be "cat", "dog", "father", "car", "ski", "cross-country", and the like, and these content features may be manually set by a relevant person, or may be obtained by inputting the information to be presented into a deep learning model and predicting the information by the model. The present embodiment does not specifically limit the structure and type of the model, and is not described herein in more detail.

The method comprises the steps of obtaining a plurality of displayed information in advance, taking content characteristics of the information and crowd classification results of corresponding audiences as training samples, taking actual watching duration of the audiences as sample labels, training a duration prediction model (a first duration prediction model), and obtaining the trained first duration prediction model. The method for acquiring the actual viewing time of the viewer will be described in detail in the following embodiments.

In another example, for each audience, the content characteristics of the information to be presented, the crowd classification result of the audience, the equipment characteristics of the presentation equipment, and the coverage crowd characteristics of the presentation equipment may be input into the second duration prediction model, so as to predict the viewing duration of the audience.

The device characteristics may be the configuration of the display device itself, and the relevant characteristics of the environment, such as the screen size of the display information, the screen brightness, the radiation range, the ambient brightness, the ambient noise, whether the current peak is on/off duty, whether the current day is on work, etc. Since the device characteristics are related to the environment around the display device, and some characteristics may change with changes in the environmental conditions, the display device may periodically update the device characteristics based on the surrounding environment, for example, the device characteristics may be acquired each time new information is displayed, or the device characteristics may be acquired every 1 hour or every 2 hours, which is not particularly limited.

The crowd coverage characteristic may be a characteristic of a crowd covered by a geographical location where the display device is located, for example, the crowd coverage characteristic may be a male-female ratio, an occupation mainly facing the crowd, income mainly facing the crowd, an age mainly facing the crowd, and the like. Because the coverage crowd characteristics are related to the geographical position of the display equipment, for the display equipment of which the geographical position is not changed, the coverage crowd characteristics are not changed generally, and the coverage crowd characteristics corresponding to each display equipment can be preset and stored, and do not need to be updated as often as the equipment characteristics.

In this example, a plurality of pieces of displayed information may be obtained in advance, and the content features of the displayed information, the crowd classification result of the audience corresponding to the information, the device features, and the covered crowd features may be used as training samples, and the actual watching time lengths of the audiences are used as sample labels to train the time length prediction model (the second time length prediction model) to obtain the trained second time length prediction model.

In this step, the first duration prediction model and the second duration prediction model may be deployed locally on the display device, and the display device may input the relevant features into the duration prediction model after obtaining the relevant features, and predict the viewing duration of the audience.

Or, the duration prediction model may also be deployed at the server, and the display device may send the relevant features to the server after obtaining the relevant features, so that the server inputs the features into the duration prediction model to predict the viewing duration of the audience.

Of course, for the features that the content features of the information to be displayed, the coverage crowd features of the device, and the like do not change due to the environment where the device is located, the server may also store the features in advance, and establish the binding relationship with the display device, so that the display device does not need to upload the features, and the server may find the corresponding features based on the binding relationship, which is not particularly limited in this embodiment.

Step 310, determining the priority of the information to be displayed based on the watching duration.

In this embodiment, the information to be displayed may be classified according to types, for example, the information is taken as an advertisement, and may be classified into a marketing advertisement, a recruitment advertisement, a house property advertisement, and the like, and a value factor corresponding to each type of information may be preset, where the value factor is generally a preset interest level of the user in the type of information, that is, a value generated by the type of information to the user. Then, a value factor corresponding to each piece of information to be displayed can be determined for each piece of information, and for each audience, the priority of the information to be displayed is determined according to the crowd classification result, the value factor and the predicted watching duration of the audience.

The following is illustrated by a specific example:

assuming that two viewers, namely viewer a and viewer B, are determined from the ambient image stream, the results of crowd classification for these two viewers can be seen in table 1 below:

TABLE 1

Of course, the crowd classification result shown in table 1 above is only an exemplary illustration, and in practical applications, the crowd classification can be performed from other dimensions.

The audience population classification result can be expressed in a vector form, that is, the audience population classification result vector corresponding to the audience A can be S _a = 0.9,0.1, 0.7,0.1, 0.7, 0.2. The crowd classification result vector for viewer B may be S _b ＝{0.1,0.9,0.6,0.2,0.1,0.1,0.8,0.1,0.1}。

Assuming that the information to be displayed is information 1, which is recruitment information for the courier, a corresponding value factor score can be set for each of the above-mentioned crowd classification types, a value factor with a higher score is set for a crowd classification type matched with the information of the recruitment information for the courier, and a value factor with a lower score is set for a crowd classification type not matched with the information of the recruitment information for the courier, as shown in the following table 2:

TABLE 2

Of course, the scores of the figure of merit shown in table 2 above are merely illustrative and are not intended to limit the present disclosure.

Similarly, the figure of merit may also be expressed as a vector, and the figure of merit vector for the example shown in Table 2 above may be W ₁ ＝{1,0,0,0.1,0.7,0.4,-0.6,0.5,0.8}。

In this embodiment, the value of the information 1 to the viewer may be determined according to the value factor corresponding to the information 1.

For example, the value can be calculated using the following equation (1):

v _i ＝S _i ·W _j (1)

wherein v is _i As the value of the information j to the viewer i, S _i Vector of results of classification of the population of viewer i, W _j Is a value factor vector for information j.

Still by taking the example described in tables 1 and 2 above as an example, assuming that the viewer is viewer a, substituting the above formula can result in:

v _a ＝S _a ·W＝{0.9，0.1，0.1，0.1，0.7，0.1，0.1，0.7，0.2}·{1，0，0，0.1，0.7，0.4，-0.6，0.5，0.8}

of course, the above method for calculating the value of the information 1 to the viewer is only an exemplary embodiment of the present specification, and other methods may be adopted to calculate the value of the information 1 to the viewer in practical applications.

For example, in other embodiments, a plurality of corresponding value factors may be set for each information type, each value factor may correspond to one crowd classification type, and when determining the display effect of the target information on the audience, the corresponding value factor may be found according to the crowd classification type to which the audience belongs, where the value factor may be a numerical value or in another form, and then the value of the target information on the audience is determined based on the value factor and the watching duration of the audience.

Still taking the crowd classification results shown in table 1 above as an example, the crowd classification results of viewer a may also be "male", "20-40 years old", or "white-collar". The value factor corresponding to the target information may be determined, for example, if the value factor corresponding to "male" is 0.8, the value factor corresponding to "female" is 0.2, the value factor corresponding to "0-20 years" is 0.3, the value factor corresponding to "20-40 years" is 0.7, the value factor corresponding to "student" is 0.1, the value factor corresponding to "white collar" is 0.9, then the value factor corresponding to viewer a may be found as: 0.8, 0.7, 0.9, which may then be summed, averaged, etc. to obtain the value of the target information to the viewer.

In this embodiment, the priority score of the information 1 to the audience may be determined based on the value and the viewing duration corresponding to each audience.

For example, the priority score of the information 1 to the viewer can be calculated using the following formula (2):

r _i ＝f(v _i ，t _i ) (2)

wherein r is _i Is the priority score, t, of information j to audience i _i Is the viewing duration of viewer i. f may be of various forms, for example a first order linear equation, where r _i And t _i 、v _i Respectively, are in positive correlation.

Still by taking the example described in tables 1 and 2 as an example, assuming that the viewer is viewer a and the viewing time of the viewer is 30 seconds, the following formula is substituted to obtain:

r _a ＝f(V _a ，30)

of course, the above examples are merely illustrative, and other methods may be used to obtain the priority score of the information 1 in practical applications.

In this embodiment, for each viewer, the priority score of the information 1 corresponding to the viewer may be obtained by calculation by using the above method, and then the priority scores of the viewers are integrated, for example, the priority scores may be summed and averaged, and the integrated priority score of the information 1 is obtained. For example, the integrated priority score of the information 1 can be obtained by the following formula (3):

wherein, R is the comprehensive priority score of the information to be displayed.

In step 312, the information to be displayed with the highest priority is determined as the second information.

And step 314, displaying the second information after the first information is displayed.

In this embodiment, the information to be displayed in the information to be displayed may be sorted in the order of priority from high to low, the information to be displayed with the highest priority is used as the second information, and the second information is played after the first information is played.

And after the second information is displayed, the second information can be marked as being displayed, so that the displayed second information can not be analyzed when the next-hop information which needs to be played after the second information is played is determined, and the repeated playing of the second information can be avoided.

Of course, if the display device circularly plays a plurality of pieces of information to be displayed, the "displayed" mark of the second information may also be cancelled after all the information in the information to be displayed is played once, and the second information may be played again in the next circulation.

As can be seen from the above description, in an embodiment of the present specification, when the information display apparatus displays information, the information display apparatus may invoke a camera module to collect an environmental image stream, determine a plurality of viewers based on the environmental image stream, predict an expected viewing time of information to be displayed for each viewer based on a crowd classification result of the viewer, and determine a priority of each piece of information to be displayed according to the expected viewing time, so as to display the information to be displayed with a highest priority after the current information is displayed.

By adopting the method, the information to be displayed with higher audience interest degree can be analyzed and displayed so as to display the information matched with the audience requirements, so that the user experience can be improved, the information which is worthless to the audience can be prevented from being displayed, and the information display efficiency is improved.

In other embodiments of the present description, the duration prediction model may be retrained based on the relevant data in the process of displaying information by the display device, so as to continuously improve the performance of the duration prediction model.

After the second information is played, the display device can also determine the actual watching time length of the second information by the audience.

For example, the environmental image stream corresponding to the second information display process may be extracted from the environmental image stream acquired by the camera module, and then a plurality of viewers are determined from the environmental image stream, and the specific method for determining the viewers may refer to the foregoing embodiment, and is not described herein again. Then, for each viewer, an image frame in which the front face of the viewer appears for the first time (hereinafter referred to as a viewing start image frame) and an image frame in which the front face of the viewer appears for the last time (hereinafter referred to as a viewing end image frame) are extracted from the environment image stream, and then a time period formed by corresponding time stamps to the two image frames is used as the actual viewing duration of the viewer.

Of course, the actual viewing duration may be determined according to the eye gazing feature of the viewer, in addition to the above-mentioned face orientation feature, and the actual viewing duration may also be determined according to the eye gazing feature of the viewer, which is not particularly limited.

The process of determining the actual viewing duration of the viewer may be performed by the display device, or may be specified by the server, which is not particularly limited.

After the actual viewing time length of each viewer is determined, the display device may further send the actual viewing time length to the server, so that the server may use the actual viewing time length as a sample label to retrain the time length prediction model (the time length prediction model here may refer to the first time length prediction model in the foregoing embodiment, and may also refer to the second time length prediction model).

For example, each display device may be configured with a corresponding server, which in turn may be attributed to a central server. For example, in a Hangzhou city, all display devices in a West lake region may be configured with a server A, all display devices in an upper city region may be configured with a server B, all display devices in a lower city region may be configured with a server C, and the server A, the server B, and the server C may be configured with a central server. See table 3 below:

TABLE 3

Wherein the apparatus 1-3 predicts the viewing time duration of the viewer using a time duration prediction model a, which is configured on the server a. The devices 4-5 predict the viewing time of the viewer using a time prediction model B, which is deployed on the server B. The devices 6-8 predict the viewer's viewing time duration using a time duration prediction model C, which is deployed on the server C.

Each display device in table 3 above can display the target information, and after the display of the target information is completed, each display device can determine the audience when the target information is played in the device and the actual watching duration of the audience, and then send the watching duration to the server configured in the device, so that the server uses the actual watching duration as the sample label to retrain the duration prediction model.

For example, the device 1 may send the actual viewing duration to the server a, and the server a may retrain the model a using the actual viewing duration as a sample label to obtain a gradient of the model a, assuming that T is T _a . The server A may apply the gradient T _a And sending the data to a central server. Similarly, the device 4 may send the actual viewing duration to the server B, which may take the actual viewing duration as a sampleThe label retrains the model b to obtain the gradient of the model b, and the assumed is T _b . The server B may apply the gradient T _b And sending the data to a central server. The device 6 may send the actual viewing duration to the server C, and the server C may retrain the model C using the actual viewing duration as a sample label to obtain a gradient of the model C, assuming that T is _c . The server C may map the gradient T _c And sending the data to a central server.

The central server may obtain a composite gradient of the duration prediction model, e.g., for T, based on the gradients sent by servers a-C _a 、T _b 、T _c The average is calculated to obtain a comprehensive gradient, then the central server can send the comprehensive gradient to the servers A-C, and the servers A-C can update the model parameters by using the comprehensive gradient.

Of course, besides the sending gradient, each device may also calculate a model parameter value of the duration prediction model in the current iteration according to the gradient obtained by the device, and send the parameter value to the central server, so that the central server obtains a comprehensive parameter value based on the model parameter value sent by each device, and sends the comprehensive parameter value to each device. The method for calculating the model parameter value based on the gradient may refer to the related art, and this embodiment is not particularly limited thereto.

Due to the fact that different information display devices are located at different geographical positions and face different crowds, audiences who actually watch information are different, and the information watching durations of different audiences are different. Therefore, more accurate audience watching duration can be obtained according to the duration prediction model, and information to be displayed, which is more matched with audience requirements and interestingness, is determined to be displayed based on the watching duration, so that user experience is improved, and information display efficiency is improved.

Corresponding to the embodiment of the information display method, the specification also provides an embodiment of an information display device.

The embodiment of the display device for the information in the specification can be applied to display equipment. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for operation through the processor of the device where the software implementation is located as a logical means. In terms of hardware, as shown in fig. 4, the apparatus for displaying information in this specification is a hardware structure diagram of a device in which the apparatus is located, and besides the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 4, the apparatus in which the apparatus is located in the embodiment may also include other hardware according to an actual function of the apparatus, which is not described again.

Referring to fig. 5, the information display apparatus can be applied to the device shown in fig. 4, and includes: a calling module 510, an audience determination module 520, a classification module 530, an information determination module 540, and a presentation module 550.

The calling module 510 calls a camera module to acquire an environment image stream in a first information display process in the first information display process;

an audience determination module 520 to determine a number of audiences of the first information based on the ambient image stream;

a classifying module 530, configured to classify the audience into groups to obtain a group classification result of each audience;

the information determining module 540 is used for determining the priority of each piece of information to be displayed based on the crowd classification result of each audience, and determining second information based on the priority;

and a display module 550 for displaying the second information after the first information is displayed.

Optionally, when determining the priority of each piece of information to be presented based on the crowd classification result of each viewer, the information determining module 540:

for each piece of information to be displayed, predicting the watching duration of each audience on the information to be displayed based on the crowd classification result of each audience and the content characteristics of the information to be displayed;

and determining the priority of the information to be displayed by integrating the watching duration of each audience.

Optionally, the information determining module 540 predicts the viewing time of each audience for the information to be displayed based on the crowd classification result of each audience and the content feature of the information to be displayed:

aiming at each audience, sending the crowd classification result of the audience and the content characteristics of the information to be displayed to a server, so that the server inputs the crowd classification result of the audience and the content characteristics of the information to be displayed into a first time length prediction model to obtain the watching time length of the audience;

and receiving the watching duration sent by the server.

Optionally, the information determining module 540 further:

acquiring the equipment characteristics of the display equipment;

the information determination module predicts the watching time of each audience to the information to be displayed based on the crowd classification result of each audience and the content characteristics of the information to be displayed:

aiming at each audience, sending the crowd classification result of the audience, the content characteristics of the information to be displayed and the equipment characteristics to a server so that the server can obtain the coverage crowd characteristics of the coverage crowd of the geographical area where the display equipment is located, and inputting the crowd classification result of the audience, the content characteristics of the information to be displayed, the equipment characteristics and the coverage crowd characteristics into a second duration prediction model to obtain the watching duration of the audience;

and receiving the watching duration sent by the server.

Optionally, the information determining module 540 further:

and after the second information is displayed, determining the actual watching duration of each audience, calculating the difference between the actual watching duration of the audience and the watching duration obtained by prediction when the updating time of the duration prediction model is reached, and updating the model parameters of the duration prediction model based on the difference.

Optionally, the information determining module 540, when determining the actual viewing time of each viewer:

extracting, for each viewer, a viewing start image frame and a viewing end image frame in which the viewer is located from the ambient image stream based on a face orientation feature of the viewer;

determining an actual viewing duration of the second information by the viewer based on the time stamp of the viewing start image frame and the time stamp of the viewing end image frame.

Optionally, value factors corresponding to each type of information are preset, and when the information determining module 540 determines the priority of the information to be displayed based on the viewing duration:

determining a value factor corresponding to the type of the information to be displayed;

and aiming at each audience, determining the priority of the information to be displayed according to the crowd classification result of the audience, the value factor and the watching duration of the audience.

Optionally, when determining the priority of the information to be displayed according to the crowd classification result of the audience, the value factor, and the watching duration of the audience, the information determining module 540:

determining the value of the information to be displayed to the audience according to the crowd classification result of the audience and the value factor;

and determining the priority of the information to be displayed based on the value and the watching duration of the audience.

Optionally, the audience determination module 520:

sending the environmental image stream to a server so that the server identifies a pedestrian from the environmental image stream and determines whether the pedestrian is a viewer of the first information based on the face orientation feature of the pedestrian;

and receiving the first information sent by the server.

Optionally, the classifying module 530:

sending the environment image stream to a server, so that the server extracts an image frame corresponding to each audience from the environment image stream aiming at each audience, and inputting the image frame into a crowd classification model to obtain a crowd classification result of the audience;

and receiving the crowd classification result of the audience, which is sent by the server.

Optionally, the call timing of the camera module includes:

the remaining display duration of the first information is less than the duration threshold.

Optionally, the information to be displayed is an advertisement, and the display device includes one or more of the following: outdoor advertisement putting equipment, elevator room advertisement putting equipment, bus stop board advertisement putting equipment.

The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.

For the device embodiment, since it basically corresponds to the method embodiment, reference may be made to the partial description of the method embodiment for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in the present specification. One of ordinary skill in the art can understand and implement without inventive effort.

The systems, apparatuses, modules or units described in the above embodiments may be specifically implemented by a computer chip or an entity, or implemented by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

Corresponding to the embodiment of the display method of the information, the present specification further provides an information display device, including: a processor and a memory for storing machine executable instructions. Wherein the processor and the memory are typically interconnected by means of an internal bus. In other possible implementations, the device may also include an external interface to enable communication with other devices or components.

In this embodiment, the processor is caused to, by reading and executing machine-executable instructions stored by the memory corresponding to exposed logic of information:

and after the first information is displayed, displaying the second information.

Optionally, when determining the priority of each information to be presented based on the result of the crowd classification of each viewer, the processor is caused to:

Optionally, when the viewing duration of each audience for the information to be presented is predicted based on the crowd classification result of each audience and the content feature of the information to be presented, the processor is caused to:

and receiving the watching duration sent by the server.

Optionally, the processor is further caused to:

acquiring the equipment characteristics of the display equipment;

the predicting the watching duration of each audience to the information to be displayed based on the crowd classification result of each audience and the content characteristics of the information to be displayed comprises the following steps:

and receiving the watching duration sent by the server.

Optionally, the processor is further caused to:

and after the second information is displayed, determining the actual watching time length of each audience, calculating the difference between the actual watching time length of the audience and the watching time length obtained by prediction when the updating time of the time length prediction model is reached, and updating the model parameters of the time length prediction model based on the difference.

Optionally, in determining the actual viewing duration of each viewer, the processor is caused to:

Optionally, a value factor corresponding to each type of information is preset, and when the priority of the information to be displayed is determined based on the viewing duration, the processor is prompted to:

Optionally, when determining the priority of the information to be presented according to the crowd classification result of the audience, the value factor and the watching duration of the audience, the processor is caused to:

Optionally, in determining the number of viewers of the first information based on the ambient image stream, the processor is caused to:

sending the environmental image stream to a server, so that the server identifies a pedestrian from the environmental image stream, and determines whether the pedestrian is a spectator of the first information based on the face orientation feature of the pedestrian;

and receiving the first information sent by the server.

Optionally, when performing crowd classification on the audiences to obtain a crowd classification result of each audience, the processor is caused to:

sending the environment image stream to a server, so that the server extracts an image frame corresponding to each audience from the environment image stream for each audience, and inputs the image frame into a crowd classification model to obtain a crowd classification result of the audience;

Optionally, the call timing of the camera module includes:

In correspondence with the foregoing embodiments of the information presentation method, the present specification also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of:

and displaying the second information after the first information is displayed.

Optionally, the determining the priority of each piece of information to be displayed based on the crowd classification result of each viewer includes:

Optionally, the predicting, based on the crowd classification result of each audience and the content feature of the information to be displayed, the viewing duration of each audience to the information to be displayed includes:

and receiving the watching duration sent by the server.

Optionally, the method further includes:

acquiring the equipment characteristics of the display equipment;

and receiving the watching duration sent by the server.

Optionally, the method further includes:

Optionally, the determining the actual viewing time of each viewer includes:

extracting, for each viewer, a viewing start image frame and a viewing end image frame in which the viewer is located from the ambient image stream based on facial orientation features of the viewer;

Optionally, a value factor corresponding to each type of information is preset, and determining the priority of the information to be displayed based on the viewing duration includes:

Optionally, the determining the priority of the information to be displayed according to the crowd classification result of the audience, the value factor and the watching duration of the audience includes:

Optionally, the determining the number of viewers of the first information based on the environmental image stream includes:

and receiving the first information sent by the server.

Optionally, the performing crowd classification on the audiences to obtain a crowd classification result of each audience includes:

Optionally, the call timing of the camera module includes:

The foregoing description of specific embodiments has been presented for purposes of illustration and description. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. A display method of information is applied to display equipment, and the method comprises the following steps:

determining a number of viewers of the first information based on the stream of ambient images;

after the first information is displayed, displaying the second information;

the determining the priority of each information to be displayed based on the crowd classification result of each audience comprises the following steps:

the priority of the information to be displayed is determined by integrating the watching duration of each audience;

the method further comprises the following steps:

acquiring the equipment characteristics of the display equipment;

and receiving the watching duration sent by the server.

2. The method of claim 1, further comprising:

3. The method of claim 2, wherein determining the actual viewing duration for each viewer comprises:

4. The method according to claim 1, wherein value factors corresponding to various types of information are preset, and the determining the priority of the information to be displayed based on the viewing duration comprises:

5. The method of claim 4, wherein the determining the priority of the information to be displayed according to the crowd classification result of the audience, the value factor and the watching duration of the audience comprises:

6. The method of claim 1, the determining a number of viewers of the first information based on the ambient image stream, comprising:

and receiving the first information sent by the server.

7. The method of claim 1, wherein the crowd classification of the viewers comprises:

8. The method of claim 1, wherein the timing of the camera module call comprises:

9. The method of claim 1, wherein the information to be displayed is an advertisement, and the display device comprises one or more of the following: outdoor advertisement putting equipment, elevator room advertisement putting equipment, bus stop board advertisement putting equipment.

10. An information display device applied to display equipment, the device comprising:

the calling module is used for calling the camera module to acquire an environment image stream in the first information display process;

a viewer determination module to determine a number of viewers of the first information based on the ambient image stream;

the display module is used for displaying the second information after the first information is displayed;

the information determination module is used for determining the priority of each piece of information to be displayed based on the crowd classification result of each audience:

the information determination module further: acquiring the equipment characteristics of the display equipment;

the information determination module predicts the watching duration of each audience to the information to be displayed based on the crowd classification result of each audience and the content characteristics of the information to be displayed:

and receiving the watching duration sent by the server.

11. The apparatus of claim 10, the information determination module further to:

12. The apparatus of claim 11, wherein the information determining module, when determining the actual viewing duration of each viewer:

13. The device of claim 10, wherein a value factor corresponding to each type of information is preset, and when the information determining module determines the priority of the information to be displayed based on the viewing duration:

14. The apparatus of claim 13, wherein the information determining module, when determining the priority of the information to be displayed according to the crowd classification result of the audience, the value factor and the watching duration of the audience:

15. The apparatus of claim 10, the audience determination module to:

and receiving the first information sent by the server.

16. The apparatus of claim 10, the classification module to:

17. The apparatus of claim 10, wherein the timing of the camera module call comprises:

18. The apparatus of claim 10, wherein the information to be displayed is an advertisement, and the display device comprises one or more of the following: outdoor advertisement putting equipment, elevator room advertisement putting equipment, bus stop board advertisement putting equipment.

19. An apparatus for displaying information, comprising:

a processor;

a memory for storing machine executable instructions;

after the first information is displayed, displaying the second information;

the processor is further caused to:

acquiring the equipment characteristics of the display equipment;

and receiving the watching duration sent by the server.