Advertisement generation method and device and electronic equipment

Abstract

The embodiment of the invention provides an advertisement generation method and device and electronic equipment, in the advertisement generation method, after the electronic equipment receives advertisement data and setting information sent by terminal equipment, features of the advertisement data are extracted according to a putting scene and a putting target of an advertisement. Then, the electronic equipment determines the first putting index, adjusts the features of the advertisement data according to the feature adjustment amplitude, screens the adjusted features, and finally, the electronic equipment generates an advertisement map according to the screened features, so that the electronic equipment can send the advertisement map to the terminal equipment according to the advertisement data sent by the terminal equipment. According to the method, the advertising indexes of the advertising data under the advertising scene and the advertising target set by the user are estimated, so that the specified advertising indexes are estimated directly based on the advertising data, the advertising indexes can be improved as the direction, and an advertising map is generated; therefore, the generated advertisement map can present a better putting effect in the putting scene specified by the user.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of smart terminal technology, and in particular to an advertisement generation method, device and electronic device. Background Art

[0002] Advertising is the promotional behavior of commodity operators and / or service providers to directly or indirectly introduce the commodities being promoted or the services being provided through certain media. Before an advertisement is placed, advertisers will specify specific user actions that are valuable to their own business (for example: activation, retention, payment, wake-up, registration and / or forms, etc., collectively referred to as "placement targets") as a measure of the effectiveness of the advertisement, and the advertising platform will charge for the advertisement based on this measure. Existing creative advertising production tools can support users to interactively design advertisements based on specific templates, and also support intelligent synthesis of advertisements based on advertising materials uploaded by users. Whether the advertising creativity fits its delivery target is a key factor affecting the effectiveness of the advertisement, but existing creative advertising production tools cannot achieve flexible adaptation to multiple delivery targets. Summary of the invention

[0003] The embodiments of the present application provide an advertisement generation method, apparatus, and electronic device. The embodiments of the present application also provide a computer-readable storage medium to estimate the delivery indicators of the advertisement data under the advertisement delivery scenarios and delivery targets set by the user based on the advertisement data sent by the terminal device, thereby realizing the estimation of the specified delivery indicators directly based on the advertisement data.

[0004] In a first aspect, an embodiment of the present application provides an advertisement generation method, comprising: receiving advertisement data and setting information sent by a terminal device; wherein the setting information comprises: an advertisement delivery scenario, a delivery target and a feature adjustment range; extracting features of the advertisement data according to the advertisement delivery scenario and the delivery target; determining a first delivery indicator corresponding to the features of the advertisement data; adjusting the features of the advertisement data according to the feature adjustment range, and filtering the adjusted features according to the first delivery indicator; and generating an advertisement image according to the filtered features.

[0005] In the above-mentioned advertisement generation method, after the electronic device receives the advertisement data and setting information sent by the terminal device, it extracts the features of the above-mentioned advertisement data according to the delivery scenario and delivery target of the above-mentioned advertisement. Then, the electronic device determines the first delivery index corresponding to the features of the above-mentioned advertisement data, adjusts the features of the above-mentioned advertisement data according to the adjustment range of the features, and filters the adjusted features according to the first delivery index. Finally, the electronic device generates an advertisement image according to the features obtained by the screening, so that the electronic device can estimate the delivery index of the advertisement data under the advertisement delivery scenario and delivery target set by the user according to the advertisement data sent by the terminal device, thereby realizing the estimation of the specified delivery index directly based on the advertisement data, and then can improve the delivery index as the direction, generate an advertisement image, so that the generated advertisement image can present a better delivery effect in the delivery scenario specified by the user.

[0006] In one possible implementation, determining the first delivery indicator corresponding to the feature of the advertisement data includes: determining the first delivery indicator corresponding to the feature of the advertisement data through a feature decoding model; wherein the feature decoding model is supervised and trained based on the delivery target.

[0007] In one possible implementation, the screening of the adjusted features according to the first delivery indicator includes: determining the second delivery indicator corresponding to the adjusted features through the feature decoding model; and selecting features whose second delivery indicator is greater than the first delivery indicator from the adjusted features.

[0008] In one possible implementation, the advertising data includes an initial advertising image, and the features of the advertising data include features of the initial advertising image; generating the advertising image based on the features obtained by screening includes: mapping the features obtained by screening back to the initial advertising image, locating the image area corresponding to the features obtained by screening in the initial advertising image to generate an advertising optimization guidance map; sending the advertising optimization guidance map and a third delivery indicator corresponding to the advertising optimization guidance map to the terminal device, so that the terminal device can adjust the initial advertising image according to the advertising optimization guidance map and the third delivery indicator.

[0009] In one possible implementation, the advertising data includes advertising materials, and the advertising materials include advertising elements and advertising templates; the characteristics of the advertising data include characteristics of the advertising elements and characteristics of the advertising templates; the setting information also includes: target delivery indicators; after extracting the characteristics of the advertising data, it also includes: embedding the characteristics of the advertising elements into the characteristics of the advertising template to obtain advertising fusion characteristics; adjusting the advertising fusion characteristics according to the characteristic adjustment range.

[0010] In one possible implementation, determining the first delivery indicator corresponding to the feature of the advertising data through a feature decoding model includes: determining the first delivery indicator corresponding to the adjusted advertising fusion feature through a feature decoding model; adjusting the feature of the advertising data according to the feature adjustment range, and screening the adjusted feature according to the first delivery indicator includes: adjusting the adjusted advertising fusion feature according to the first delivery indicator and the feature adjustment range until the first delivery indicator is equal to the target delivery indicator; and obtaining the target advertising fusion feature when the first delivery indicator is equal to the target delivery indicator.

[0011] In one possible implementation manner, generating an advertisement image according to the features obtained through screening includes: generating an advertisement image according to the target advertisement fusion features.

[0012] In one possible implementation manner, after generating the advertisement image, the method further includes: sending the advertisement image to the terminal device so that the terminal device displays the advertisement image.

[0013] In a second aspect, an embodiment of the present application provides an advertisement generating device, which is included in an electronic device, and has the function of implementing the electronic device behavior in the first aspect and the possible implementation of the first aspect. The function can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules or units corresponding to the above functions. For example, a receiving module, an extraction module, a determination module, an adjustment module, a screening module, and a generation module.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; a memory; multiple applications; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the electronic device, enable the electronic device to execute the method provided in the first aspect.

[0015] It should be understood that the second and third aspects of the embodiments of the present application are consistent with the technical solutions of the first aspect of the embodiments of the present application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar and will not be repeated here.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer executes the method provided in the first aspect.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program, which, when executed by a computer, is used to execute the method provided in the first aspect.

[0018] In one possible design, the program in the fifth aspect may be stored in whole or in part on a storage medium packaged together with the processor, or may be stored in whole or in part on a memory not packaged together with the processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application;

[0020] Figure 2 A schematic diagram of a system architecture provided for an embodiment of the present application;

[0021] Figure 3 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0022] Figure 4 A schematic diagram of an application scenario provided for another embodiment of the present application;

[0023] Figure 5 A schematic diagram of an application scenario provided for yet another embodiment of the present application;

[0024] Figure 6 Flowchart of an advertisement generation method provided for one embodiment of the present application

[0025] Figure 7 A schematic diagram of setting parameters related to delivery index estimation provided by an embodiment of the present application;

[0026] Figure 8 A schematic diagram of inputting an advertisement picture provided by an embodiment of the present application;

[0027] Fig. 9 A schematic diagram of a terminal sending an advertisement picture and related settings provided in one embodiment of the present application;

[0028] Fig.10 A schematic diagram of advertisement image feature extraction provided by one embodiment of the present application;

[0029] Fig.11 A schematic diagram of target delivery index estimation provided by one embodiment of the present application;

[0030] Fig.12 A schematic diagram of a server / cloud sending a delivery index estimation result to a terminal provided by an embodiment of the present application;

[0031] Fig.13 A schematic diagram of a display of a delivery indicator estimation result and an interactive interface provided by an embodiment of the present application;

[0032] Fig.14A flowchart of an advertisement generation method provided by another embodiment of the present application;

[0033] Fig.15 A schematic diagram of setting advertisement generation related parameters provided by an embodiment of the present application;

[0034] Fig.16 A schematic diagram of synthesizing / adjusting an advertisement image provided by an embodiment of the present application;

[0035] Fig.17 A schematic diagram of uploading advertisement images and related settings provided for one embodiment of the present application;

[0036] Fig.18 A schematic diagram of extracting advertisement image features provided by an embodiment of the present application;

[0037] Fig.19 A schematic diagram of target delivery index estimation provided by one embodiment of the present application;

[0038] Fig. 20 A schematic diagram of fine-tuning image features provided by an embodiment of the present application;

[0039] Fig.21 A schematic diagram of generating a creative layout optimization guide provided by one embodiment of the present application;

[0040] Fig. 22 A schematic diagram of a user adjusting a creative layout according to a creative optimization guide provided in one embodiment of the present application;

[0041] Fig.23 A flowchart of an advertisement generation method provided in yet another embodiment of the present application;

[0042] Fig.24 A schematic diagram of setting advertisement generation related parameters provided by an embodiment of the present application;

[0043] Fig.25 A schematic diagram of inputting advertising materials and templates provided for one embodiment of the present application;

[0044] Fig.26 A schematic diagram of setting a target delivery indicator provided by an embodiment of the present application;

[0045] Fig. 27 A schematic diagram of a terminal uploading advertising materials and related settings to a server / cloud provided in one embodiment of the present application;

[0046] Fig.28 A schematic diagram of advertisement material feature extraction provided by one embodiment of the present application;

[0047] Fig.29A schematic diagram of advertising material feature fusion provided by an embodiment of the present application;

[0048] Fig.30 A schematic diagram of fusion feature fine-tuning and advantageous feature screening provided in one embodiment of the present application;

[0049] Fig.31 A schematic diagram of generating an advertisement image based on advantageous features provided by an embodiment of the present application;

[0050] Fig.32 A schematic diagram of sending an advertisement image provided by an embodiment of the present application;

[0051] Fig.33 A flowchart of an advertisement generation method provided in yet another embodiment of the present application;

[0052] Fig.34 A schematic diagram of the structure of an electronic device provided in another embodiment of the present application;

[0053] Fig.35 A schematic diagram of the structure of an electronic device provided in yet another embodiment of the present application. DETAILED DESCRIPTION

[0054] The terms used in the implementation section of this application are only used to explain the specific embodiments of this application and are not intended to limit this application.

[0055] In an advertisement generation scheme provided by the existing related technology, the advertisement generation platform combines intelligent advertisement production capabilities and can automatically generate advertisement creatives without relying on templates. Users add local product main pictures or fill in product detail page links in the material upload interface, upload logos and / or slogans and other supplementary materials; select marketing goals, such as event promotions, new products and / or holiday promotions; set the advertisement placement location, such as the opening screen interface of the application software and / or the advertisement slot of the video stream; select the advertisement size and click the creative generation button, and the platform will intelligently synthesize the uploaded advertisement materials into a finished advertisement.

[0056] However, in this solution, ad generation cannot flexibly adapt to various advertising materials, delivery scenarios and delivery targets; and it cannot automatically generate ads based on delivery indicators in specific delivery scenarios.

[0057] Another advertising generation solution provided by the existing related technology is to interactively generate advertisements based on templates. Users set the template type, industry and / or image size and other items in the template selection interface according to the delivery requirements to filter the required templates. Then, the user selects the template, uploads the advertising material, and edits the main and subtitles, buttons and / or description text on the template editing page; edits the main image, background image, text bottom frame and / or button bottom frame and other image materials; adjusts the position and / or transparency of the text and image. After a series of interactive adjustments, the creative advertisement production is completed.

[0058] However, the advertisement generation in this scheme lacks targeted design guidance, and the degree of fit between the advertisement design and the delivery target is difficult to estimate.

[0059] Based on the above problems, an embodiment of the present application provides an advertisement generation method, which is an interactive creative advertisement generation method guided by delivery targets. It can estimate the delivery indicators of the advertisement data under the advertisement delivery scenarios and delivery targets set by the user based on the advertisement data sent by the terminal device, thereby directly estimating the specified delivery indicators based on the advertisement data, and making the generated advertisements adaptable to multiple delivery targets.

[0060] The advertisement generation method provided in the embodiment of the present application can be applied to an electronic device, wherein the electronic device can be a server, the server can be a server with an advertisement generation function, and the server can be set in the cloud; the embodiment of the present application does not impose any restrictions on the specific type of the electronic device.

[0061] For example, Figure 1 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application is shown as follows: Figure 1 As shown, the electronic device 100 includes a processor 110 and a communication interface 120. Optionally, the electronic device 100 may also include a memory 130. The processor 110, the communication interface 120 and the memory 130 may communicate with each other through an internal connection path to transmit control and / or data signals, the memory 130 is used to store a computer program, and the processor 110 is used to call and run the computer program from the memory 130.

[0062] The processor 110 and the memory 130 may be combined into a processing device, or more commonly, they are independent components, and the processor 110 is used to execute program codes stored in the memory 130. In specific implementation, the memory 130 may also be integrated into the processor 110, or may be independent of the processor 110.

[0063] In addition, in order to make the functions of the electronic device 100 more complete, the electronic device 100 may further include one or more of an input unit 160 and a display unit 170 .

[0064] Optionally, the electronic device 100 may further include a power supply 150 for providing power to various devices or circuits in the electronic device 100 .

[0065] It should be understood that Figure 1 The processor 110 in the electronic device 100 shown may be a system on chip SOC, and the processor 110 may include a central processing unit (CPU), and may further include other types of processors, such as a graphics processing unit (GPU) and the like.

[0066] For ease of understanding, the following embodiments of the present application will be described in detail with Figure 1 Taking the electronic device with the structure shown as an example, the advertisement generation method provided in the embodiment of the present application is specifically described in combination with the accompanying drawings and application scenarios.

[0067] Figure 2 A schematic diagram of a system architecture provided for an embodiment of the present application is shown in FIG. Figure 2 As shown, the system architecture of this embodiment may include: a terminal and a server / cloud. The terminal is mainly used to support image display, image editing and data communication, and the server / cloud is mainly used for image feature extraction and fine-tuning, delivery index estimation and data communication.

[0068] In specific implementation, the terminal device may be a terminal device with a display screen, such as a mobile phone, a tablet computer or a personal computer (PC). The terminal device may generally support a variety of applications, such as: a camera program, an image management program, an image processing program, a word processing program, a telephone application, an email program, an instant messaging program, a network communication program, a media player program, a geographic location positioning program or a time management program.

[0069] The server / cloud device may be a Windows server or other server that can provide simultaneous access to multiple devices, or may be a server cluster consisting of multiple regions, multiple computer rooms, and / or multiple servers. The server / cloud device generally supports message storage and distribution programs, large-scale data storage programs, large-scale data processing programs, and / or data redundancy backup programs, etc. In the embodiment of the present application, the server / cloud device may be the electronic device 100.

[0070] The terminal part can include the following five functional modules:

[0071] ①Candidate material display and interactive interface: used to display advertising materials and support users to select materials.

[0072] ② Advertising image display and interactive interface: used to display advertising images and creative layout optimization guidance, supporting users to add, delete, move, zoom in and out of advertising materials.

[0073] ③Display and interactive interface of delivery indicator estimation results: used to display the delivery indicator estimation results, and supports users to edit delivery indicators for the system to reverse the advertising creative layout based on the indicators.

[0074] ④ Function / parameter setting interface: supports users to set advertising delivery scenarios, delivery targets, function selection (for example: selecting from indicator estimation function, interactive advertising generation function and automated advertising generation function), and feature adjustment range in interactive / automated advertising generation.

[0075] ⑤Information upload module: used to upload advertising images, upload user-set delivery scenarios, delivery targets, selected system functions, feature adjustment ranges, etc.

[0076] ⑥Information download module: used to download calculated delivery indicators and layout optimization guidance.

[0077] The server / cloud mainly has the following four capability modules:

[0078] ⑦Image feature extraction: Extract the semantic features of advertising images based on the delivery indicator estimation model.

[0079] ⑧ Delivery indicator estimation: Use the feature decoding model trained based on the specified delivery target supervision to estimate the delivery indicator corresponding to the image features.

[0080] ⑨ Fine-tuning of image features and screening of advantageous features: Randomly fine-tune the features of the advertising image, and use the feature decoding model trained based on the specified delivery target to infer the delivery indicators corresponding to the fine-tuned features. Feature fine-tuning is repeated N times, and feature screening involves three scenarios: 1) Select the feature that makes the target delivery indicator the highest and higher than the original indicator among the N features; 2) Select the first feature that makes the target delivery indicator reach K times the original indicator; 3) If N fine-tunings do not obtain features that are better than the original features, the user is prompted that the set feature adjustment range may need to be increased. Among them, the feature adjustment range is used to indicate the position translation step and / or scale scaling degree of the feature in the advertising image.

[0081] ⑩Creative layout optimization guidance generation: Map the fine-tuned segment features back to the original image and locate the image area corresponding to the segment features. Mark the area in the advertisement image using heat intensity and / or detection boxes to guide creative layout optimization.

[0082] Figure 3 A schematic diagram of an application scenario provided by an embodiment of the present application, Figure 3The application scenario shown is to estimate the designated delivery index based on the input advertisement image. Specifically, the user loads the advertisement image into the advertisement delivery tool or creative generation tool of the terminal device and sets the delivery scenario and delivery target, and estimates the designated delivery index based on the input advertisement image and delivery scenario. Figure 3 For example, the user enters the ad Figure 1 , set the delivery target to activation rate, delivery scenario to A, and the system outputs ads Figure 1 The activation rate in scenario A is 3.2% (upper left). The ad image and delivery scenario remain unchanged, but the delivery target is changed to wake-up rate. The system outputs the ad. Figure 1 The wake-up rate in scenario A is 4.5% (upper right). The ad image and delivery target remain unchanged. The delivery scenario is changed to scenario B. The system outputs the ad. Figure 1 The activation rate in scenario B is 3.7% (lower left). The delivery target and delivery scenario remain unchanged, but the ad image is changed. The system outputs the ad. Figure 2 The activation rate in scenario A is 2.1% (lower right figure).

[0083] Figure 4 A schematic diagram of an application scenario provided by another embodiment of the present application, Figure 4 The application scenario shown is interactive creative layout optimization aimed at improving the target delivery index. After the user sets or adjusts the creative layout of the advertisement, the system estimates the delivery index based on the current layout and makes layout optimization suggestions in the form of heat maps or rectangular boxes. The user and the system interactively optimize the creative layout of the advertisement. Figure 4 For example, in the first step, the user enters the advertisement Figure 1 .0, set the delivery target to activation rate, delivery scenario to scenario A, and the system estimates the advertising Figure 1 .0 has an activation rate of 2.2% in scene A, and provides layout optimization suggestions in the form of a rectangular frame, indicating that the main body of the picture needs to be moved to the position of the rectangular frame and scaled to the size of the rectangular frame; in the second step, the user updates the advertisement image to 1.1 based on the layout optimization suggestions, and the system estimates that the advertisement Figure 1 .1 has an activation rate of 2.9% in scene A, and provides layout optimization suggestions in the form of a rectangular frame; in the third step, the user continues to update the ad image to 1.2 based on the layout optimization suggestions, and the system estimates that the ad Figure 1 .2 The activation rate in scene A is 3.5%, and layout optimization suggestions are provided in a rectangular frame. Users can choose to save the ad after completing any step.

[0084] Figure 5 A schematic diagram of an application scenario provided in yet another embodiment of the present application, Figure 5 The application scenario shown is to automatically generate advertisements based on the delivery indicators preset by the user. Figure 5For example, in step 1, the user inputs advertising materials into the terminal, sets the delivery scenario to scenario A, the delivery target to activation rate, and the target delivery index to 3.2%. The terminal uploads the advertising materials and related settings to the server / cloud. In step 2, the system randomly combines the advertising materials into advertising images, extracts image features, and estimates delivery indicators. From step 3 to step K-1, the system fine-tunes the image features, progressively updates the image features in the direction of improving the target delivery index, and iterates multiple times until the estimated delivery index of the image reaches the user's preset value. In step K, the system maps the features covering the target delivery index back to the original image to generate an advertising image.

[0085] It should be noted that the “system” mentioned in the description of the embodiments of the present application can be understood as a system running on the electronic device 100 .

[0086] The delivery scenarios covered by the embodiments of the present application may be shown in Table 1.

[0087] Table 1

[0088]

[0089] The delivery targets covered by the embodiments of the present application may be as shown in Table 2.

[0090] Table 2

[0091]

[0092]

[0093] The system functions covered by the embodiments of the present application may be as shown in Table 3.

[0094] Table 3

[0095] System Features Functional Description Delivery indicator estimation Estimate the delivery indicators of ads in a specified delivery scenario based on images Interactive ad generation Guide users to optimize creative layouts in order to improve designated delivery targets Automatic ad generation The system automatically generates advertising ideas that meet the user's preset indicators

[0096] The following introduces the advertisement generation method provided in the embodiment of the present application in combination with the system architecture and application scenarios.

[0097] Figure 6 This is a flowchart of an advertisement generation method provided by an embodiment of the present application. This embodiment takes the opening screen scene as an example to show Figure 3 The process in the application scenario shown in Figure 1 is as follows. Figure 6 As shown, the above advertisement generation method may include:

[0098] Step 601, function / parameter setting: In the terminal function / parameter setting interface, set the delivery scenario to screen opening, the delivery target to wake-up, and the system function to index estimation.

[0099] Figure 7 A schematic diagram of setting parameters related to delivery index estimation provided by an embodiment of the present application is shown in FIG. Figure 7 As shown, the terminal may provide a function / parameter setting interface. In the function / parameter setting interface, the user sets the delivery scenario to screen opening, the delivery target to wake-up, and the system function to index estimation.

[0100] Step 602: input advertisement picture.

[0101] See also Figure 8 , this embodiment provides two advertising image input methods: a) the user directly loads the advertising image to be evaluated into the terminal, and b) the user combines the advertising material and template into an advertising image at the terminal. Figure 8 A schematic diagram of inputting an advertisement image provided in accordance with an embodiment of the present application.

[0102] Step 603, upload the advertisement image and related settings.

[0103] See also Fig. 9 , the terminal sends the advertising image and the user-entered settings (delivery scenario: screen opening, delivery target: wake-up, system function: delivery indicator estimation) to the server / cloud. Fig. 9 A schematic diagram of a terminal sending an advertisement image and related settings provided in one embodiment of the present application.

[0104] Step 604: extract advertisement image features.

[0105] like Fig.10 As shown in the figure, the server / cloud extracts image features based on the delivery scene (screen opening) and delivery target (wake-up) set by the user. Fig.10 A schematic diagram of advertisement image feature extraction provided in one embodiment of the present application.

[0106] Step 605: estimate the target delivery index.

[0107] See also Fig.11 ,The server / cloud estimates the delivery indicator (e.g. 3.2% wake-up rate) corresponding to the image features through a feature decoding model ,trained based on the supervision of the specified delivery target (wake-up rate). Fig.11 A schematic diagram of target delivery indicator estimation provided for one embodiment of the present application.

[0108] Step 606, sending the delivery index estimation result.

[0109] See also Fig.12 , the server / cloud sends the estimated results of the delivery indicators to the terminal, Fig.12 A schematic diagram of a server / cloud sending a delivery indicator estimation result to a terminal provided in an embodiment of the present application.

[0110] Step 607: The user views the estimated results of the delivery index.

[0111] See also Fig.13 , the user can view the estimated wake-up rate of the advertisement in the terminal's delivery indicator estimation result display and interactive interface. The presentation method of the delivery indicator estimation result may include one or a combination of the following: graded numbers, percentage numbers, graded texts, and graphics (color bar + arrow + graded numbers, color bar + arrow + percentage numbers, or color bar + arrow + graded text), etc. Of course, the above presentation methods are only examples, and the terminal can also use other presentation methods, which are not limited in the embodiments of the present application. Fig.13 A schematic diagram of the display of the delivery indicator estimation results and the interactive interface provided for one embodiment of the present application.

[0112] The above-mentioned advertisement generation method can estimate the target delivery index of the advertisement delivery scenario set by the user based on the advertisement image. The existing related technology reflects the delivery effect of the template through the historical delivery data of the advertisement template. The synthesized advertisement cannot flexibly adapt to various advertising materials, delivery scenarios and delivery targets, and does not support direct quantitative estimation of the delivery effect of the advertisement. The above-mentioned advertisement generation method can estimate the target delivery index of the advertisement delivery scenario set by the user based on the input advertisement image, and the estimation of the delivery index is more targeted and flexible.

[0113] Fig.14 A flowchart of an advertisement generation method provided in another embodiment of the present application, which shows Figure 4 The process in the application scenario shown in Figure 1 is as follows. Fig.14 As shown, the above advertisement generation method may include:

[0114] Step 1401, setting parameters related to advertisement generation.

[0115] Specifically, see Fig.15 The terminal provides a function / parameter setting interface, in which the user can set the delivery scenario to open screen, the delivery target to wake-up, the system function to interactive advertisement generation, and the feature adjustment amplitude to H. Among them, H indicates the position translation step length and / or scale scaling degree of the feature in the advertisement image, and H is proportional to the creative optimization guidance generation delay. Fig.15 A schematic diagram of setting advertisement generation related parameters provided in one embodiment of the present application.

[0116] Step 1402, synthesize / adjust the advertisement image.

[0117] Specifically, see Fig.16 , the user can select advertising elements and put them into the picture display and interactive interface on the terminal's candidate material display and interactive interface, and perform operations such as translation, zooming in and / or reducing the advertising elements. Fig.16 A schematic diagram of synthesizing / adjusting an advertisement image provided by one embodiment of the present application.

[0118] Step 1403, upload the advertisement image and related settings.

[0119] Specifically, see Fig.17 , the terminal sends the advertising image and the user-entered settings (delivery scenario: screen opening; delivery target: wake-up; system function: interactive advertising generation; feature adjustment range: H) to the server / cloud. Fig.17 A schematic diagram of uploading advertisement images and related settings provided for one embodiment of the present application.

[0120] Step 1404: extract advertisement image features.

[0121] Specifically, see Fig.18 The server / cloud extracts image features based on the delivery scenario (screen opening) and delivery target (wake-up) set by the user. Fig.18 A schematic diagram of extracting advertisement image features provided in accordance with an embodiment of the present application.

[0122] Step 1405, target delivery index estimation.

[0123] Specifically, see Fig.19 ,The server / cloud estimates the delivery indicator (e.g. 3.2% wake-up rate) corresponding to the image features through a feature decoding model ,trained based on the supervision of the specified delivery target (wake-up rate). Fig.19 A schematic diagram of target delivery indicator estimation provided for one embodiment of the present application.

[0124] Step 1406, fine-tune the image features.

[0125] Specifically, see Fig. 20 The server / cloud fine-tunes the image features according to the feature adjustment amplitude H, where H is the sum of the changes in multiple dimensions, for example: the local feature translation step (pixel) + the local feature enlargement / reduction amplitude (pixel). Fig. 20 , the original feature vector is [1,2,3,4,5,6,7,6,5,4,3,2,1], the local feature [6,7,6] is enlarged by 2 pixels, and the fine-tuned feature is [1,2,3,4,6,6.5,7,6.5,6,4,3,2,1]; the local feature [5,6,7,6,5] of the original feature is reduced by 1 pixel, and the fine-tuned feature is [1,2,3,4,4.75,5.5,6.5,5.5,4.75,4,3,2,1]; the local feature [6,7,6] of the original feature is translated to the right by 2 pixels, and the fine-tuned feature is [1,2,3,4,5,5.33,5.67,6,7,6,3,2,1]. The holes caused by feature translation and reduction can be filled by methods such as nearest neighbor interpolation. Fig. 20 A schematic diagram of fine-tuning image features provided in accordance with an embodiment of the present application.

[0126] Step 1407, screening of advantageous features.

[0127] Specifically, feature screening involves the following three scenarios: 1) The server performs N feature fine-tuning and selects the feature among the N features that makes the target delivery index the highest and higher than the original delivery index; 2) Selects the first feature that makes the target delivery index reach K times the original index, where K can be a system built-in parameter or a user-set parameter; 3) If N fine-tunings fail to obtain a feature that is better than the original feature, the user is prompted that the set feature adjustment range may need to be increased.

[0128] Step 1408: Generate creative layout optimization guidance.

[0129] Specifically, see Fig.21 , the server maps the segment features that are fine-tuned relative to the original image features among the filtered dominant features back to the original image, and locates the image area corresponding to the segment features. Fig.21 A schematic diagram of the creative layout optimization guidance generation provided in one embodiment of the present application.

[0130] Step 1409, sending the delivery index estimation results and creative layout optimization guidance.

[0131] Specifically, the server / cloud sends the delivery indicator estimation results and creative layout optimization guidance to the terminal.

[0132] Step 1410: The user views the estimated results of the delivery indicators.

[0133] Specifically, see Fig. 22 The terminal can display the estimated results of the delivery indicators and the interactive interface. The user can view the estimated wake-up rate of the advertisement in the terminal delivery indicator estimated results display and the interactive interface, view the creative layout optimization guidance in the terminal advertisement image display and the interactive interface (mark the guidance in the advertisement image by means of heat intensity and / or detection frame, etc.), and adjust the advertisement image according to the guidance. Fig. 22 A schematic diagram of a user adjusting a creative layout according to creative optimization guidance provided in one embodiment of the present application.

[0134] Then, steps 1402 to 1410 are repeated to perform progressive optimization of the creative layout guided by the delivery target.

[0135] The above-mentioned advertisement generation method provides a method for progressive optimization of creative layout guided by delivery target. In the process of advertisement design, the goal of improving the delivery target set by the user is taken as the direction, and creative layout optimization suggestions are progressively provided in combination with delivery scenarios and advertisement materials. The user follows the system suggestions to gradually optimize the advertisement layout, so that the advertisement can obtain better delivery effect in the specified delivery scenario.

[0136] Fig.23A flowchart of an advertisement generation method provided in another embodiment of the present application. This embodiment shows Figure 5 The process in the application scenario shown in Figure 1 is as follows. Fig.23 As shown, the above advertisement generation method may include:

[0137] Step 2301, setting parameters related to advertisement generation.

[0138] Specifically, see Fig.24 The terminal can display a function / parameter setting interface. The user can set the delivery scene to open screen, the delivery target to wake-up, the system function to automatic advertisement generation, and the feature adjustment range to H in the function / parameter setting interface of the terminal. Fig.24 A schematic diagram of setting advertisement generation related parameters provided in one embodiment of the present application.

[0139] Step 2302, input advertising materials and templates.

[0140] Specifically, see Fig.25 The terminal can display an advertising picture display and interactive interface, and the user can input advertising materials into the advertising picture display and interactive interface of the terminal. Fig.25 A schematic diagram of inputting advertising materials and templates provided for one embodiment of the present application.

[0141] Step 2303, setting the target delivery indicator.

[0142] Specifically, see Fig.26 , the user can input the target wake-up rate indicator (e.g., 3.5%) into the terminal's delivery indicator estimation result and interactive interface. Fig.26 A schematic diagram of setting target delivery indicators provided for one embodiment of the present application.

[0143] Step 2304, upload the advertising material and related settings.

[0144] Specifically, see Fig. 27 , the terminal sends the advertising material and the user input settings (delivery scenario: screen opening, delivery target: wake-up, system function: automatic advertising generation, feature adjustment range: H, target delivery index: 3.5%) to the server / cloud. Fig. 27 A schematic diagram of a terminal uploading advertising materials and related settings to a server / cloud provided in one embodiment of the present application.

[0145] Step 2305: extracting advertising material features.

[0146] Specifically, see Fig.28 The server / cloud extracts material features according to the delivery scenario (screen opening) and delivery target (wake-up) set by the user, where the advertising material includes advertising elements and advertising templates. Fig.28A schematic diagram of advertisement material feature extraction provided in accordance with an embodiment of the present application.

[0147] Step 2306: Advertisement material feature integration.

[0148] Specifically, the server / cloud embeds the advertising element features into the advertising template features, such as Fig.29 As shown, the advertisement template features are [1,2,3,4,5,6,7], and an advertisement element feature is [8,8,8]. After embedding the element into the template, the fused features can be [1,8,8,8,5,6,7] or [1,2,8,8,8,6,7], etc. (the element features directly cover the template feature fragments). Fig.29 A schematic diagram of the fusion of advertising material features provided in accordance with an embodiment of the present application.

[0149] Step 2307, fusion feature fine-tuning and dominant feature screening.

[0150] Specifically, see Fig.30 The server / cloud fine-tunes the fusion feature within a certain range (feature adjustment range H) and infers the corresponding delivery index (for example, 2.9%) based on the fine-tuned feature. Each fine-tuning does not necessarily improve the template delivery index. When the inferred index reaches the user's preset index (3.5%), the feature screening is completed and the feature is the dominant feature. Fig.30 A schematic diagram of fusion feature fine-tuning and advantageous feature screening provided for one embodiment of the present application.

[0151] Step 2308, generating a creative layout that meets the indicators based on the advantageous features and generating an advertisement image.

[0152] Specifically, Fig.31 As shown, the server / cloud can generate an advertising image based on the advantageous features and use the feature extraction model to perform inverse operations. Fig.31 A schematic diagram of generating an advertisement image based on advantageous features provided in one embodiment of the present application.

[0153] Step 2309, sending the advertisement image.

[0154] Specifically, see Fig.32 , the server / cloud sends the generated advertising image to the terminal. Fig.32 A schematic diagram of sending an advertisement image provided by an embodiment of the present application.

[0155] Step 2310: The user views the generated advertisement image.

[0156] Specifically, the user can view the advertisement image on the advertisement image display and interactive interface of the terminal. Afterwards, if the user sets a new target delivery index, steps 2303 to 2310 can be repeated to perform layout optimization for the target delivery index newly set by the user.

[0157] The above-mentioned advertisement generation method provides a layout optimization method for preset delivery indicators. In the process of advertisement design, the target delivery indicator set by the user is used as the direction, and the delivery scenario and advertisement material are combined to find a creative layout that meets the set indicator. The user can modify the target delivery indicator multiple times to obtain an advertisement image that can achieve the estimated value of the corresponding delivery indicator.

[0158] As described above, an embodiment of the present application provides an advertisement generation method, which uses a feature extraction model based on supervised training of advertising delivery targets to extract semantic features of advertisement images, and uses a feature decoding model based on supervised training of corresponding delivery targets to infer delivery indicators corresponding to the image features, thereby enabling estimation of specified delivery indicators directly based on advertisement images.

[0159] Fig.33 A flowchart of an advertisement generation method provided in yet another embodiment of the present application is as follows: Fig.33 As shown, the above advertisement generation method may include:

[0160] Step 3301: The electronic device 100 receives advertisement data and setting information sent by a terminal device.

[0161] The above setting information may include: the advertising delivery scenario, delivery target and feature adjustment amplitude. The feature adjustment amplitude is used to indicate the position translation step and / or scale scaling degree of the feature in the advertising image. In other words, the feature adjustment amplitude may be the sum of the changes in multiple dimensions, for example: the position translation step (pixel) of the local feature + the amplitude (pixel) of the local feature enlargement / reduction.

[0162] Step 3302: The electronic device 100 extracts features of the advertisement data according to the advertisement delivery scenario and delivery target.

[0163] In a specific implementation, the electronic device 100 may use a feature extraction model based on supervised training of advertising delivery targets to extract features of the above-mentioned advertising data.

[0164] Step 3303: The electronic device 100 determines a first delivery indicator corresponding to the characteristics of the above-mentioned advertising data.

[0165] Specifically, the electronic device 100 determines the first delivery indicator corresponding to the feature of the above-mentioned advertising data as follows: the electronic device 100 determines the first delivery indicator corresponding to the feature of the above-mentioned advertising data through a feature decoding model; wherein the above-mentioned feature decoding model is based on the above-mentioned delivery target supervision training.

[0166] In step 3304, the electronic device 100 adjusts the characteristics of the advertisement data according to the characteristic adjustment range, and screens the adjusted characteristics according to the first delivery indicator.

[0167] Specifically, screening the adjusted features according to the first delivery index may be: determining the second delivery index corresponding to the adjusted features through the feature decoding model, and selecting features whose second delivery index is greater than the first delivery index from the adjusted features.

[0168] Step 3305: The electronic device 100 generates an advertisement image according to the features obtained through screening.

[0169] In one implementation of this embodiment, the above-mentioned advertising data may include an initial advertising image, and the characteristics of the above-mentioned advertising data include the characteristics of the initial advertising image; in this way, the electronic device 100 generates an advertising image based on the characteristics obtained by screening: the electronic device 100 maps the characteristics obtained by screening back to the above-mentioned initial advertising image, and locates the image area corresponding to the above-mentioned characteristics obtained by screening in the above-mentioned initial advertising image to generate an advertising optimization guidance map; then, the electronic device 100 sends the above-mentioned advertising optimization guidance map and the third delivery indicator corresponding to the above-mentioned advertising optimization guidance map to the terminal device, so that the terminal device can adjust the above-mentioned initial advertising image according to the above-mentioned advertising optimization guidance map and the third delivery indicator.

[0170] In another implementation of this embodiment, the above-mentioned advertising data may include advertising materials, and the above-mentioned advertising materials include advertising elements and advertising templates; the characteristics of the above-mentioned advertising data include characteristics of advertising elements and characteristics of advertising templates; the above-mentioned setting information may also include: target delivery indicators; thus, in step 3302, after extracting the characteristics of the above-mentioned advertising data, the electronic device 100 may also embed the characteristics of the above-mentioned advertising elements into the characteristics of the above-mentioned advertising template to obtain advertising fusion characteristics; then, the above-mentioned advertising fusion characteristics are adjusted according to the above-mentioned characteristic adjustment range.

[0171] In this way, step 3303 can be: the electronic device 100 determines the first delivery indicator corresponding to the adjusted advertising fusion feature through the feature decoding model; step 3304 can be: the electronic device 100 adjusts the adjusted advertising fusion feature according to the above-mentioned first delivery indicator and the above-mentioned feature adjustment range until the first delivery indicator is equal to the target delivery indicator; and obtains the target advertising fusion feature when the first delivery indicator is equal to the target delivery indicator.

[0172] Therefore, step 3305 may be: generating an advertisement image according to the target advertisement fusion feature. Further, after generating the advertisement image, the electronic device 100 may also send the advertisement image to the terminal device so that the terminal device displays the advertisement image.

[0173] In the above-mentioned advertisement generation method, after the electronic device 100 receives the advertisement data and setting information sent by the terminal device, it extracts the features of the above-mentioned advertisement data according to the delivery scenario and delivery target of the above-mentioned advertisement. Then, the electronic device 100 determines the first delivery index corresponding to the features of the above-mentioned advertisement data, adjusts the features of the above-mentioned advertisement data according to the adjustment range of the features, and filters the adjusted features according to the first delivery index. Finally, the electronic device 100 generates an advertisement image according to the features obtained by the screening, so that the electronic device 100 can estimate the delivery index of the advertisement data under the advertisement delivery scenario and delivery target set by the user according to the advertisement data sent by the terminal device, thereby realizing the estimation of the specified delivery index directly based on the advertisement data, and then can improve the delivery index as the direction, generate an advertisement image, so that the generated advertisement image can present a better delivery effect in the delivery scenario specified by the user.

[0174] It is to be understood that some or all of the steps or operations in the above embodiments are merely examples, and the present application embodiments may also perform other operations or variations of various operations. In addition, the various steps may be performed in different orders presented in the above embodiments, and it is possible that not all of the operations in the above embodiments need to be performed.

[0175] It is understandable that, in order to realize the above functions, the electronic device includes hardware and / or software modules corresponding to the execution of each function. In combination with the algorithm steps of each example described in the embodiments disclosed in this application, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application in combination with the embodiments, but such implementation should not be considered to exceed the scope of this application.

[0176] This embodiment can divide the electronic device into functional modules according to the above method embodiment. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0177] Fig.34A schematic diagram of the structure of an electronic device provided in another embodiment of the present application, in which each functional module is divided according to each function, Fig.34 A possible schematic diagram of the composition of the electronic device 3400 involved in the above embodiment is shown. Fig.34 As shown, the electronic device 3400 may include: a receiving module 3401, an extracting module 3402, a determining module 3403, an adjusting module 3404, a screening module 3405 and a generating module 3406;

[0178] The receiving module 3401 is used to receive the advertisement data and setting information sent by the terminal device; wherein the setting information may include: advertisement delivery scene, delivery target and feature adjustment range;

[0179] Extraction module 3402, used to extract features of the advertisement data according to the advertisement delivery scenario and delivery target;

[0180] The determination module 3403 is used to determine the first delivery indicator corresponding to the feature of the above-mentioned advertisement data; specifically, the determination module 3403 is specifically used to determine the first delivery indicator corresponding to the feature of the above-mentioned advertisement data through a feature decoding model; wherein the above-mentioned feature decoding model is based on the above-mentioned delivery target supervision training;

[0181] An adjustment module 3404 is used to adjust the characteristics of the advertisement data according to the characteristic adjustment range;

[0182] A screening module 3405 is used to screen the features adjusted by the adjustment module 3404 according to the first delivery index;

[0183] The generating module 3406 is used to generate an advertisement image according to the features obtained by filtering the filtering module 3405 .

[0184] It should be noted that this application Fig.33 All relevant contents of each step involved in the illustrated method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0185] The electronic device 3400 provided in this embodiment is used to execute the present application Fig.33 The advertisement generating method provided by the illustrated embodiment can therefore achieve the same effect as the above-mentioned method.

[0186] Fig.35 A schematic diagram of the structure of an electronic device provided in another embodiment of the present application, Fig.34 The electronic device shown is different in that Fig.35In the electronic device 3400 shown, the screening module 3405 is specifically used to determine the second delivery index corresponding to the adjusted feature through the above-mentioned feature decoding model; and select the feature whose second delivery index is greater than the first delivery index from the adjusted features.

[0187] In one implementation of this embodiment, the above-mentioned advertising data includes an initial advertising image, and the characteristics of the above-mentioned advertising data include the characteristics of the initial advertising image; thus, the generation module 3406 is specifically used to map the filtered features back to the above-mentioned initial advertising image, and locate the image area corresponding to the filtered features in the above-mentioned initial advertising image to generate an advertising optimization guidance map.

[0188] Furthermore, the electronic device 3400 may further include: a sending module 3407;

[0189] The sending module 3407 is used to send the above-mentioned advertising optimization guidance map and the third delivery indicator corresponding to the above-mentioned advertising optimization guidance map to the terminal device, so that the terminal device can adjust the above-mentioned initial advertising map according to the above-mentioned advertising optimization guidance map and the above-mentioned third delivery indicator.

[0190] In another implementation of this embodiment, the above-mentioned advertisement data includes advertisement materials, and the above-mentioned advertisement materials include advertisement elements and advertisement templates; the characteristics of the above-mentioned advertisement data include characteristics of advertisement elements and characteristics of advertisement templates; the above-mentioned setting information also includes: target delivery indicators; thus, the electronic device 3400 may further include: an embedding module 3408;

[0191] An embedding module 3408 is used to embed the features of the advertisement element into the features of the advertisement template after the features of the advertisement data are extracted by the extraction module 3402, so as to obtain advertisement fusion features;

[0192] The adjustment module 3404 is specifically configured to adjust the advertisement fusion feature according to the feature adjustment range.

[0193] The determination module 3403 is specifically configured to determine the first delivery index corresponding to the adjusted advertisement fusion feature through a feature decoding model;

[0194] The adjustment module 3404 is further configured to adjust the adjusted advertisement fusion feature according to the first delivery index and the feature adjustment range until the first delivery index is equal to the target delivery index;

[0195] The screening module 3405 is specifically used to obtain the target advertisement fusion feature when the first delivery index is equal to the target delivery index.

[0196] The generating module 3406 is specifically used to generate an advertisement image according to the above-mentioned target advertisement fusion characteristics.

[0197] Furthermore, the electronic device 3400 may further include: a sending module 3407;

[0198] The sending module 3407 is used to send the advertisement image generated by the generating module 3406 to the terminal device so that the terminal device can display the advertisement image.

[0199] It should be noted that this application Fig.33 All relevant contents of each step involved in the illustrated method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0200] The electronic device 3400 provided in this embodiment is used to execute the present application Fig.33 The advertisement generating method provided by the illustrated embodiment can therefore achieve the same effect as the above-mentioned method.

[0201] It should be understood that the electronic device 3400 may correspond to Figure 1 The electronic device 100 shown in FIG. The functions of the receiving module 3401 and the sending module 3407 can be Figure 1 The processor 110 and the communication interface 120 in the electronic device 100 are implemented; the functions of the extraction module 3402, the determination module 3403, the adjustment module 3404, the screening module 3405, the generation module 3406 and the embedding module 3408 can be implemented by Figure 1 The processor 110 in the electronic device 100 is shown implemented.

[0202] In the case of adopting an integrated unit, the electronic device 3400 may include a processing module, a storage module, and a communication module.

[0203] The processing module can be used to control and manage the actions of the electronic device 3400, for example, it can be used to support the electronic device 3400 to execute the steps executed by the above modules. The storage module can be used to support the electronic device 3400 to store program codes and data, etc. The communication module can be used to support the communication between the electronic device 3400 and other devices.

[0204] Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logic boxes, modules and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, and so on. The storage module can be a memory. The communication module can specifically be a device that interacts with other electronic devices, such as a radio frequency circuit, a Bluetooth chip and / or a Wi-Fi chip.

[0205] In one embodiment, when the processing module is a processor and the storage module is a memory, the electronic device 3400 involved in this embodiment can be a Figure 1 Device of the structure shown.

[0206] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer executes the present application. Fig.33 The method provided by the illustrated embodiment.

[0207] The present application also provides a computer program product, which includes a computer program, which, when executed on a computer, enables the computer to execute the present application. Fig.33 The method provided by the illustrated embodiment.

[0208] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0209] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0210] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0211] In several embodiments provided in the present application, any function can be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0212] The above is only a specific implementation of the present application. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. The protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An advertisement generating method, characterized in that: include: Receiving advertisement data and setting information sent by a terminal device; wherein the setting information includes: advertisement delivery scenario, delivery target and feature adjustment range; Extracting features of the advertisement data according to the advertisement delivery scenario and delivery target; Determine a first delivery indicator corresponding to the feature of the advertisement data; Adjusting the characteristics of the advertisement data according to the characteristic adjustment range, and screening the adjusted characteristics according to the first delivery indicator; Generate an advertisement graph based on the features obtained by screening.

2. The method according to claim 1, characterized in that The determining of the first delivery indicator corresponding to the feature of the advertisement data includes: A first delivery indicator corresponding to the feature of the advertisement data is determined through a feature decoding model; wherein the feature decoding model is trained based on the delivery target supervision.

3. The method according to claim 1, characterized in that The screening of the adjusted features according to the first delivery indicator includes: Determining the second delivery indicator corresponding to the adjusted feature through the feature decoding model; Select a feature in which the second delivery index is greater than the first delivery index from the adjusted features.

4. The method according to claim 1, characterized in that: The advertisement data includes an initial advertisement image, and the characteristics of the advertisement data include characteristics of the initial advertisement image; Generating an advertisement graph according to the features obtained by screening includes: Mapping the features obtained by screening back to the initial advertisement image, and locating the image area corresponding to the features obtained by screening in the initial advertisement image to generate an advertisement optimization guidance map; The advertisement optimization guidance map and a third delivery indicator corresponding to the advertisement optimization guidance map are sent to the terminal device, so that the terminal device can adjust the initial advertisement map according to the advertisement optimization guidance map and the third delivery indicator.

5. The method according to claim 1, characterized in that The advertisement data includes advertisement materials, and the advertisement materials include advertisement elements and advertisement templates; the characteristics of the advertisement data include characteristics of advertisement elements and characteristics of advertisement templates; The setting information also includes: target delivery indicators; After extracting the features of the advertisement data, the method further includes: Embedding the features of the advertisement element into the features of the advertisement template to obtain advertisement fusion features; The advertisement fusion feature is adjusted according to the feature adjustment range.

6. The method according to claim 5, characterized in that The determining of the first delivery indicator corresponding to the feature of the advertisement data includes: Determining a first delivery indicator corresponding to the adjusted advertisement fusion feature through a feature decoding model; Adjusting the feature of the advertisement data according to the feature adjustment range, and screening the adjusted feature according to the first delivery indicator includes: Adjusting the adjusted advertisement fusion feature according to the first delivery indicator and the feature adjustment range until the first delivery indicator is equal to the target delivery indicator; A target advertisement fusion feature is obtained when the first delivery indicator is equal to the target delivery indicator.

7. The method according to claim 6, characterized in that Generating an advertisement graph according to the features obtained by screening includes: An advertisement image is generated according to the target advertisement fusion feature.

8. The method according to claim 7, characterized in that After the advertisement image is generated, the method further includes: The advertisement image is sent to the terminal device so that the terminal device displays the advertisement image.

9. An advertisement generating device, characterized in that: include: A receiving module, configured to receive advertisement data and setting information sent by a terminal device; wherein the setting information includes: advertisement delivery scenario, delivery target, and feature adjustment range; An extraction module, used to extract features of the advertisement data according to the advertisement delivery scenario and delivery target; A determination module, used to determine a first delivery indicator corresponding to the feature of the advertisement data; An adjustment module, configured to adjust the characteristics of the advertisement data according to the characteristic adjustment range; A screening module, used for screening the features adjusted by the adjustment module according to the first delivery index; A generating module is used to generate an advertisement image according to the features obtained by screening by the screening module.

10. An electronic device, characterized in that: include: one or more processors; Memory; Multiple applications; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the electronic device, enable the electronic device to perform the method as described in any one of claims 1-8.

Images