Automatic material generation method and device, equipment and storage medium

Abstract

The invention relates to an automatic material generation method and device, equipment and a storage medium, and the method comprises the steps: obtaining a to-be-added target material and a target cue word used for generating a picture, the target cue word comprises target region description information of a target region, and the target region does not comprise key content in the to-be-generated picture; calling a preset picture generation model, and obtaining a reference picture including the target area by using the target cue word; according to a mark box generation strategy, marking the target area to obtain a target picture comprising at least one mark box; and according to a mark box in the target picture, adding the target material into the mark box to obtain the target material. According to the method, the picture not including key content in the designated area can be automatically generated, and the designated material is added to the designated area to obtain the advertisement material, that is, the method can automatically generate the advertisement material.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular, to a method, device, equipment, and storage medium for automatically generating materials. Background Art

[0002] With the rapid development of technology, the technology of automatically generating advertising pictures has become increasingly popular, setting off a new wave in the advertising industry. As the core driving force in this field, AIGC (Artificial Intelligence Generated Content) technology, with its powerful generation ability, is widely used in producing high-quality advertising picture materials and has become an indispensable part of the advertising production process.

[0003] Currently, when generating advertising picture materials with the help of AIGC technology, the main process is as follows: First, use AIGC technology to generate a reference picture according to the preset picture content, and then technicians determine the material filling area in the reference picture and fill the corresponding material into this area to obtain the required materials.

[0004] However, in actual applications, some scenarios have strict position requirements for material filling and need to fill the material into a fixed position. Once there is key content in this fixed position, the filling of this material will interfere with the overall presentation effect of the picture. Facing this problem, currently technicians have to intervene manually, screening one by one among the numerous generated reference pictures to find those pictures where there is no key content in the fixed position, and then using these screened pictures to generate materials with the materials. This process not only consumes a lot of time and energy of technicians, but also seriously reduces the generation efficiency of the required materials. Therefore, there is an urgent need for a method for automatically generating materials. Summary of the Invention

[0005] This application provides a method, device, equipment, and storage medium for automatically generating materials. This method can automatically generate a picture that does not include key content in a specified area, and add a specified material to this specified area to obtain advertising materials, that is, this method can automatically generate advertising materials.

[0006] In a first aspect, this application provides a method for automatically generating materials, and the method includes:

[0007] Obtain a target material to be added and a target prompt word for generating a picture, where the target prompt word includes target area description information of a target area, and the target area does not include key content in the picture to be generated;

[0008] Call a preset picture generation model, and use the target prompt word to obtain a reference picture including the target area;

[0009] Mark the target area according to a marking box generation strategy to obtain a target picture including at least one marking box;

[0010] Add the target material into the marking frame in the target picture to obtain the target material.

[0011] Optionally, marking the target area according to the marking frame generation strategy to obtain a target picture including at least one marking frame includes:

[0012] Obtain the association relationship between the area and the area description information;

[0013] Determine the target area on the target picture according to the target area description information and the association relationship;

[0014] Use the marking frame to mark the target area to obtain a target picture including at least one marking frame.

[0015] Optionally, when the area description information includes the area description sub-information corresponding to each target material, adding the target material into the marking frame in the target picture to obtain the target material includes:

[0016] Obtain the area information corresponding to each marking frame;

[0017] Determine the marking frame corresponding to each target material according to the area description sub-information of each target material and the area information corresponding to each marking frame;

[0018] Add each target material into the corresponding marking frame to obtain the target material.

[0019] Optionally, the method further includes:

[0020] Display the target picture including at least one marking frame;

[0021] Receive the modification information of the marking frame in the target picture by the user;

[0022] Modify the marking frame in the target picture according to the modification information to obtain a target picture including the modified one.

[0023] Optionally, the target prompt word further includes content requirements. Invoking a preset picture generation model and using the target prompt word to obtain a reference picture including the target area includes:

[0024] Obtain the content requirements in the target prompt word;

[0025] Invoke the picture generation model and use the content requirements and the area description information to obtain a reference picture that meets the content requirements and the area description information.

[0026] Optionally, the method further includes:

[0027] On the initial picture, determine a reference area;

[0028] Add a description to the reference area to obtain reference area description information;

[0029] According to the reference area and the reference area description information, establish an association relationship between the area and the area description information.

[0030] Optionally, the method further includes:

[0031] Detect whether the marking box meets a preset condition on the target picture;

[0032] When the marking box does not meet the preset condition on the target picture, process the target picture based on a preset picture processing strategy.

[0033] In a second aspect, the present application provides a material automatic generation device, and the device includes:

[0034] An acquisition unit, configured to acquire a target material to be added and a target prompt word for generating a picture, where the target prompt word includes target area description information of a target area, and the target area does not include key content in the picture to be generated;

[0035] A calling unit, configured to call a preset picture generation model and use the target prompt word to obtain a reference picture including the target area;

[0036] A marking unit, configured to mark the target area according to a marking box generation strategy to obtain a target picture including at least one marking box;

[0037] An adding unit, configured to add the target material into the marking box according to the marking box in the target picture to obtain a target material.

[0038] Optionally, the device further includes a strategy unit, and the strategy unit is configured to:

[0039] Obtain an association relationship between an area and area description information;

[0040] According to the target area description information and the association relationship, determine the target area on the target picture;

[0041] Use the marking box to mark the target area to obtain a target picture including at least one marking box.

[0042] Optionally, when the area description information includes area description sub-information corresponding to each target material, the adding unit is configured to:

[0043] Obtain the region information corresponding to each bounding box;

[0044] Determine the bounding box corresponding to each target material according to the region description sub-information of each target material and the region information corresponding to each bounding box;

[0045] Add each target material to the corresponding bounding box to obtain the target material.

[0046] Optionally, the device further includes a display unit, and the display unit is used for:

[0047] Display a target image including at least one bounding box;

[0048] Receive the modification information of the bounding box in the target image by the user;

[0049] Modify the bounding box in the target image according to the modification information to obtain a target image including the modified one.

[0050] Optionally, the target prompt further includes content requirements, and the calling unit is used for:

[0051] Obtain the content requirements in the target prompt;

[0052] Call an image generation model, and use the content requirements and the region description information to obtain a reference image that meets the content requirements and the region description information.

[0053] Optionally, the device further includes a creation unit, and the creation unit is used for:

[0054] Determine a reference region on the initial image;

[0055] Add a description to the reference region to obtain reference region description information;

[0056] Establish an association relationship between the region and the region description information according to the reference region and the reference region description information.

[0057] Optionally, the device further includes a detection unit, and the detection unit is used for:

[0058] Detect whether the bounding box meets a preset condition on the target image;

[0059] When the bounding box does not meet the preset condition on the target image, process the target image based on a preset image processing strategy.

[0060] In a third aspect, the present application provides a material automatic generation device, including: at least one communication interface; at least one bus connected to the at least one communication interface; at least one processor connected to the at least one bus; at least one memory connected to the at least one bus, wherein the processor is configured to:

[0061] Obtain a target material to be added and a target prompt word for generating a picture, the target prompt word includes target area description information of a target area, and the target area does not include key content in the picture to be generated;

[0062] Call a preset picture generation model, and use the target prompt word to obtain a reference picture including the target area;

[0063] According to a marking box generation strategy, mark the target area to obtain a target picture including at least one marking box;

[0064] According to the marking box in the target picture, add the target material into the marking box to obtain a target material.

[0065] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned material automatic generation method is implemented.

[0066] The above technical solutions provided by the embodiments of the present application have the following advantages compared with the prior art:

[0067] In the embodiment of the present application, a target material to be added and a target prompt word for generating a picture are obtained, the target prompt word includes target area description information of a target area, and the target area does not include key content in the picture to be generated; call a preset picture generation model, and use the target prompt word to obtain a reference picture including the target area; use a preset strategy to mark the target area to obtain a target picture including a marking box; according to the marking box on the target picture, add the target material into the marking box to obtain a target material. It can be seen that the present application can automatically generate a picture that does not include key content in a specified area, and automatically add a specified material to the specified area to obtain an advertising material, so that the specified material in the advertising material does not cover the key content in the picture. In summary, the present application can automatically generate advertising materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] The drawings here are incorporated into the description and form a part of this description, showing embodiments consistent with the present invention and used together with the description to explain the principles of the present invention.

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0070] One or more embodiments are exemplarily illustrated by the pictures in the corresponding drawings. These exemplary illustrations do not limit the embodiments. Elements with the same reference numerals in the drawings represent similar elements. Unless otherwise stated, the drawings in the figures do not constitute a scale limitation.

[0071] Figure 1 It is a schematic flowchart of a method for automatically generating materials provided by an embodiment of the present application;

[0072] Figure 2 It is a schematic diagram of a reference picture provided by an embodiment of the present application;

[0073] Figure 3 It is a schematic flowchart of a method for marking a marking box provided by an embodiment of the present application;

[0074] Figure 4 It is a schematic flowchart of a method for determining target materials provided by an embodiment of the present application;

[0075] Figure 5 It is a schematic flowchart of another method for modifying a marking box provided by an embodiment of the present application;

[0076] Figure 6 It is a schematic flowchart of a method for generating a reference picture provided by an embodiment of the present application;

[0077] Figure 7 It is a schematic flowchart of another method for creating a relationship provided by an embodiment of the present application;

[0078] Figure 8 It is a flowchart of a method for automatically generating materials provided by an embodiment of the present application

[0079] Figure 9 It is a schematic flowchart of a device for automatically generating materials provided by an embodiment of the present application;

[0080] Figure 10 It is a schematic diagram of a device for automatically generating materials provided by an embodiment of the present application. Detailed implementation manners

[0081] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0082] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0083] With the rapid development of science and technology, the technology of automatically generating advertising pictures has become more and more popular, setting off a new wave in the advertising industry. As the core driving force in this field, AIGC (artificial intelligence generated content) technology is widely used to produce high-quality advertising picture materials with its powerful generation capabilities, becoming an indispensable part of the advertising production process. At present, the generation of advertising picture materials with the help of AIGC technology mainly follows the following process: first, AIGC technology is used to generate reference pictures according to the preset picture content, and then the technicians determine the material filling area in the reference picture and fill the corresponding materials into the area to obtain the required materials. However, in actual applications, some scenes have strict position requirements for material filling, and the materials need to be filled into a fixed position. Once there is key content in the fixed position, the filling of the material will interfere with the overall presentation of the picture. Faced with this problem, technicians currently have to intervene manually, screen one by one in the many reference pictures generated, find those pictures where there is no key content in the fixed position, and then use these screened pictures and materials to generate materials. This process not only consumes a lot of time and energy of technicians, but also seriously reduces the generation efficiency of the required materials.

[0084] In order to solve the above problems, the embodiment of the present application provides a material automatic generation method, which can automatically generate a picture that does not include key content in a specified area, and add a specified material to the specified area to obtain an advertising material, that is, the method can automatically generate an advertising material, such as Figure 1 As shown, the specific steps include:

[0085] Step 101, obtaining a target material to be added and a target prompt word for generating a picture.

[0086] Among them, the target prompt includes the target area description information of the target area, and the target area description information includes at least the area description information of each material's location. The area description information can be the approximate location of the material. For example, the commodity logo is located in the upper right corner. Of course, the area description information can also be the central coordinates of the material, the shape and size of the area where it is located. For example, the central coordinates are XX, the shape is circular, and the radius is XX. The target area does not include the key content in the picture to be generated, that is to say, the target area only includes the non-key content in the picture. The key content refers to the elements that play a key role in conveying the core information in the picture, such as people, main objects, etc. The non-key content is some relatively secondary elements that play an auxiliary or foil role, such as the background without any decoration. As Figure 2 shown, if the key content in the picture to be generated is a cat, then the non-key content is the area corresponding to the area other than the cat.

[0087] In this step, the technician can determine the target material to be added according to the business requirements. At the same time, the technician can also generate a target prompt according to the area where the material needs to be set. For example, when the material needs to be set in the upper right corner, the target prompt is that the material is located in the upper right corner. Or, the terminal is provided with a prompt generation page, which includes a blank picture of the same size as the reference picture. The technician can draw the required area in the blank picture. The terminal receives the area, and generates the target prompt of the picture according to the coordinates, shape and size of the area.

[0088] It should be noted that the execution subject of the embodiments of the present application can be a terminal, and the terminal can be a device such as a computer, a notebook, a smart phone, etc., or a server, which is not limited here. The embodiments of the present application are described by taking the execution subject as a terminal as an example.

[0089] Step 102, call a preset picture generation model, and use the target prompt to obtain a reference picture including the target area.

[0090] Among them, the preset picture generation model is a pre-trained machine learning model, and this machine learning model is trained using a large amount of data, and is used to identify the input prompt and output the picture corresponding to the prompt. The target area in the reference picture does not include the key content, that is, the target area only includes the non-key content. For example, the non-key content is the picture background, and the target area of the reference picture only includes the background content.

[0091] In this step, the target prompt is input into the preset picture generation model to obtain a reference picture.

[0092] Step 103, mark the target area according to the marking box generation strategy to obtain a target picture including at least one marking box.

[0093] Among them, the target picture is composed of a marking frame and a reference picture.

[0094] In this step, the marking frame generation strategy is used to determine the target area in the reference picture. The specific steps are as follows: according to the target area description information in the target prompt, find the area corresponding to the marking frame in the reference picture, and this area is the target area. Then, use the marking frame to mark this target area to obtain a target picture including the marking frame.

[0095] The specific steps to determine the target area can be as follows: preset the association relationship between the area description information and the area. Then, according to the target area description information and this association relationship, determine the area corresponding to the marking frame in the reference picture, and this area is the target area. When the target area description information includes the coordinates, shape, and size of the material center, the position of the marking frame corresponding to the material can also be directly determined according to this information, and the area corresponding to it in the reference picture is the target area.

[0096] Step 104, add the target material to the marking frame on the target picture to obtain the target material.

[0097] Among them, the target material is composed of the reference picture and the target material, that is to say, the target material does not include the marking frame.

[0098] In this step, on the terminal, display the target picture and the marking frame on the target picture. At this time, the user can add the target material to the marking frame. The display page also includes a confirmation button. When the user clicks this confirmation button, the target material is generated according to the target picture and the target material.

[0099] In the embodiment of the present application, obtain the target material to be added and the target prompt for generating the picture. The target prompt includes the target area description information of the target area, and the key content in the picture to be generated is not included in the target area; call the preset picture generation model, and use the target prompt to obtain a reference picture including the target area; use the preset strategy to mark the target area to obtain a target picture including the marking frame; according to the marking frame on the target picture, add the target material to the marking frame to obtain the target material. It can be seen that the present application can automatically generate a picture that does not include key content in the specified area, and automatically add the specified material to the specified area to obtain advertising material, so that the specified material in the advertising material does not cover the key content in the picture. In summary, the present application can automatically generate advertising material.

[0100] In the embodiments of the present application, after obtaining the reference picture, it is also necessary to analyze the region description information to determine the target region in the reference picture, and then use a marking frame to mark the target region to obtain the target picture. Therefore, the embodiments of the present application provide a marking frame marking method, which defines the marking frame generation strategy in step 103, such as Figure 3 as shown, and the specific steps are as follows:

[0101] Step 301, obtain the association relationship between the region and the region description information.

[0102] For example, the region is the lower left corner region, and its corresponding region description information is the lower left corner.

[0103] Step 302, determine the target region on the target picture according to the target region description information and the association relationship.

[0104] In this step, the target region description information is compared with the region description information in this association relationship. When they match, the region corresponding to the matching region description information is determined as the target region on the target picture.

[0105] For example, if the region description information is the lower right corner, the lower right corner in the target picture is determined as the target region.

[0106] Step 302, use a marking frame to mark the target region to obtain a target picture including at least one marking frame.

[0107] In this step, use a marking frame to mark the target region. After all the target regions are marked, a target picture including at least one marking frame is obtained.

[0108] It should be noted that the size of the marking frame is the same as the size of the target region. In this way, after determining the target region, the size of the target region is determined, and then the size of the marking frame is set according to it.

[0109] In the embodiments of the present application, when the region description information includes region description sub-information corresponding to each target material, the picture generation model will generate a reference picture including multiple target regions according to these region description sub-information. Then, the target regions in the reference picture are marked to obtain a target picture including multiple marking frames. During the marking process, the region description information corresponding to each marking frame is also stored. In this way, the region description information corresponding to each marking frame can be compared with the region description sub-information corresponding to each target material to determine the marking frame corresponding to each target material. Finally, each target material can be added to the corresponding marking frame to obtain the target material. Therefore, the embodiments of the present application provide a target material determination method, which is Figure 4 as shown, a further limitation of step 103, and the specific steps include:

[0110] Step 401: Obtain the region information corresponding to each bounding box.

[0111] In this step, during the bounding process of the Figure 2 described embodiment, the region information corresponding to each bounding box is also stored.

[0112] Step 402: Determine the bounding box corresponding to each target material according to the region descriptor information of each target material and the region information corresponding to each bounding box.

[0113] In this step, for each target material, compare the region descriptor information of the target material with the region information corresponding to each bounding box. When the region descriptor information matches the region information corresponding to a certain bounding box, it indicates that the target material corresponds to this bounding box, and the bounding box corresponding to the target material is obtained. By executing the above method, the bounding box corresponding to each target material is obtained.

[0114] Step 403: Add each target material into the corresponding bounding box to obtain the target material.

[0115] In this step, for each target material, determine the bounding box corresponding to the target material and add the target material into this bounding box. When all target materials are successfully added, the target material is obtained.

[0116] It should be noted that after adding the target material to the bounding box, the size of the target material in the target material can be set to the original size of the target material, or the size of the target material in the target material can be set according to the size of the bounding box.

[0117] Of course, the size of the material in the material may not meet the user's requirements. A material size adjustment page can also be displayed, and then on this page, the size of the material in the target material can be adjusted, so that the target material better meets the user's requirements.

[0118] In the embodiment of the present application, when the bounding box in the target image is too large or too small, the size and position of the bounding box can also be adjusted, so that the modified bounding box better meets the user's requirements. Therefore, the embodiment of the present application provides a method for modifying a bounding box, and this method is as Figure 5 shown, and the specific steps include:

[0119] Step 501: Display a target image including at least one bounding box.

[0120] In this step, after obtaining the target image, the target image and the bounding box on the target image can also be displayed to enable the user to modify the position and size of the bounding box.

[0121] Step 502: Receive the modification information of the bounding box in the target picture from the user.

[0122] The modification information includes the position coordinates, size, etc. of the modified bounding box.

[0123] In this step, a target picture confirmation page is displayed on the terminal. There is a modification button on this page. When the user clicks this modification button, they can enter the bounding box modification page to modify the position and size of the bounding box on this page. This display page also includes a confirmation button. When the user clicks the confirmation button, the user can add materials to the target picture to obtain the target material.

[0124] Step 503: Modify the bounding box in the target picture according to the modification information to obtain a target picture including the modified bounding box.

[0125] In this step, modify the position coordinates in the bounding box according to the position coordinates in the modification information to obtain a target picture including the modified bounding box.

[0126] Since the modified bounding box meets the user's requirements, the size of the corresponding material in the target material can be set according to the size of the bounding box.

[0127] In the embodiment of the present application, in order to generate a reference picture of specified content, the target indicator further includes content requirements. In this way, after inputting the target prompt into the picture generation model, a reference picture that meets the content requirements and region description information can be generated. Therefore, the embodiment of the present application provides a method for generating a reference picture, and the method is as Figure 6 shown. The specific steps include:

[0128] Step 601: Obtain the content requirements in the target prompt.

[0129] The content requirements are the content that the picture to be generated needs to generate. For example, the picture to be generated includes a blue background and a yellow cat.

[0130] In this step, extract the region description information and content requirements from the target prompt and input these data into the picture generation model.

[0131] Step 602: Call the picture generation model and use the content requirements and region description information to obtain a reference picture that meets the content requirements and region description information.

[0132] In this step, input the content requirements and region description information into the picture input generation model to obtain a reference picture that meets the content requirements and region description information.

[0133] It should be noted that after obtaining the reference picture, it is also possible to detect whether the reference picture meets the content requirements and whether it meets the area description information. When the reference picture does not meet the content requirements or / and the reference picture does not meet the area description information, it is also possible to check whether the reference picture can be modified. If it cannot be modified, the reference picture can be discarded.

[0134] In the embodiments of the present application, on the initial picture, a reference area is determined, and reference area description information is added to the reference area. Then, according to the reference area and the reference area description information, an association relationship between the area and the area description information is created. Therefore, the embodiments of the present application also include a relationship creation method, as Figure 7 shown, and the specific steps include:

[0135] Step 701, on the initial picture, determine the reference area.

[0136] Among them, the size of the initial picture is the same as that of the reference picture, or when the initial picture is enlarged or reduced by a preset ratio, the size-changed initial picture is the same as that of the reference picture.

[0137] In this step, the initial prompt word can be input into the preset picture generation model to obtain the initial picture, where the initial prompt word is set by the technician according to experience. Or the technician searches for a picture with the same size as the reference picture on the website. It is also possible to obtain the initial picture based on other methods, which is not limited here.

[0138] After that, the technician can, according to the requirements, mark at least one reference area on the initial picture.

[0139] Step 702, add a description to the reference area to obtain the reference area description information.

[0140] In this step, the technician can add a description to each reference area to obtain the reference area description information corresponding to each reference area.

[0141] Step 703, according to the reference area and the reference area description information, establish an association relationship between the area and the area description information.

[0142] In this step, an association relationship between the area and the area description information can be established according to each reference area and its corresponding reference area description information.

[0143] In the embodiments of the present application, after obtaining the target image, it is also possible to detect whether the size and position of the bounding box in the target image meet the requirements. When they do not meet the requirements, they are discarded or otherwise processed to ensure that the obtained target image meets the requirements. Therefore, the embodiments of the present application provide a method for detecting bounding boxes. The specific steps of this method include: detecting whether the bounding box meets the preset conditions on the target image; when the bounding box does not meet the preset conditions on the target image, processing the target image based on the preset image processing strategy.

[0144] Among them, the preset condition is that the area corresponding to the bounding box does not overlap with the area where the key content is located. For example, the key content in the target image is a cat, and the area of the bounding box does not overlap with the area where the cat is located. The preset image processing strategy can be to discard the image or modify the bounding box in the image to make it meet the preset conditions.

[0145] In this step, since the target prompt includes the image content, which is actually the key content in the target image, it is possible to analyze this image content, and then find the area corresponding to the key content in the target image, and then detect whether the area corresponding to the key content overlaps with the area corresponding to the bounding box. When the two overlap, it is determined that the bounding box does not meet the preset conditions on the target image, and the target image needs to be processed based on the preset image processing strategy.

[0146] Furthermore, in this step, a content recognition model and an image recognition model can be used to analyze the image content, and then find the area corresponding to the key content in the target image. It is also possible for technicians to analyze the image content and then find the area corresponding to the key content in the target image, which is not limited here.

[0147] As Figure 8 shown, the embodiments of the present application present a process of an automatic material generation method, which focuses on AIGC (Artificial Intelligence Generated Content) image production and related material processing. The specific steps are as follows:

[0148] First, the user enters an initial Prompt (prompt) in the material generation system. The initial Prompt includes content requirements, which are used to indicate the generation of images that meet the content requirements. After the entry is completed, the system displays a blank page, and the user can check the preset blank areas on it, which are used for the locations of the materials in the subsequent images.

[0149] Next, the system further constructs a new Prompt based on the initial Prompt entered by the user and the selected area. After the new Prompt is generated, it is immediately input into the AIGC model, and then an AIGC image is generated. After the image is generated, the system automatically adds a bounding box to it, which is used to identify the area where the target material is located.

[0150] Subsequently, the AIGC image with a marked box is displayed on the front-end interface for users to view. In this step, users can manually correct the marked box on the displayed image according to actual needs to ensure the accuracy of its position and scope. The AIGC image with a marked box that has been corrected or does not need to be corrected (if manually confirmed to meet the requirements) becomes the finally used image.

[0151] Finally, add the target material to the marked box to complete the generation of the target material.

[0152] As Figure 9 shown, the embodiment of the present application provides a device for automatically generating materials. This device corresponds to the method embodiment and specifically includes:

[0153] An acquisition unit 901, configured to acquire a target material to be added and a target prompt word for generating an image. The target prompt word includes target area description information of the target area, and the target area does not include key content in the image to be generated;

[0154] An invocation unit 902, configured to invoke a preset image generation model and use the target prompt word to obtain a reference image including the target area;

[0155] A marking unit 903, configured to mark the target area according to a marking box generation strategy to obtain a target image including at least one marking box;

[0156] An addition unit 904, configured to add the target material to the marking box according to the marking box in the target image to obtain a target material.

[0157] Optionally, the device further includes a strategy unit 905, and the strategy unit 905 is configured to:

[0158] Obtain the association relationship between the area and the area description information;

[0159] Determine the target area on the target image according to the target area description information and the association relationship;

[0160] Use the marking box to mark the target area to obtain a target image including at least one marking box.

[0161] Optionally, when the area description information includes area description sub-information corresponding to each target material, the addition unit 906 is configured to:

[0162] Obtain the area information corresponding to each marking box;

[0163] Determine the marking box corresponding to each target material according to the area description sub-information of each target material and the area information corresponding to each marking box;

[0164] Add each target material to the corresponding marked box to obtain the target material.

[0165] Optionally, the device further includes a display unit 907, and the display unit 907 is configured to:

[0166] Display a target picture including at least one marked box;

[0167] Receive modification information of the marked box in the target picture from the user;

[0168] Modify the marked box in the target picture according to the modification information to obtain a target picture including the modified one.

[0169] Optionally, the target prompt word further includes content requirements, and the calling unit 902 is configured to:

[0170] Obtain the content requirements in the target prompt word;

[0171] Call an image generation model, and use the content requirements and the region description information to obtain a reference picture that meets the content requirements and the region description information.

[0172] Optionally, the device further includes a establishing unit 908, and the establishing unit 908 is configured to:

[0173] Determine a reference region on the initial picture;

[0174] Add a description to the reference region to obtain reference region description information;

[0175] Establish an association relationship between the region and the region description information according to the reference region and the reference region description information.

[0176] Optionally, the device further includes a detecting unit 909, and the detecting unit 909 is configured to:

[0177] Detect whether the marked box meets a preset condition on the target picture;

[0178] When the marked box does not meet the preset condition on the target picture, process the target picture based on a preset picture processing strategy.

[0179] As Figure 10 shown, an embodiment of the present application provides a material automatic generation device, including a processor 1001, a communication interface 1002, a memory 1003, and a communication bus 1004. Among them, the processor 1001, the communication interface 1002, and the memory 1003 communicate with each other through the communication bus 1004.

[0180] A memory 1003 for storing computer programs;

[0181] In an embodiment of the present application, when the processor 1001 executes the program stored on the memory 1003, it implements the material automatic generation method provided in any of the foregoing method embodiments, including:

[0182] Obtain a target material to be added and a target prompt for generating a picture, where the target prompt includes target area description information of a target area, and the target area does not include key content in the picture to be generated;

[0183] Call a preset picture generation model, and use the target prompt to obtain a reference picture including the target area;

[0184] According to a marking box generation strategy, mark the target area to obtain a target picture including at least one marking box;

[0185] According to the marking box in the target picture, add the target material into the marking box to obtain a target material.

[0186] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps executed by the material automatic generation method provided in any of the foregoing method embodiments.

[0187] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0188] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the related technology can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0189] It should be understood that the terms used herein are for the purpose of describing particular example embodiments only and are not intended to be limiting. Unless the context clearly dictates otherwise, the singular forms "a", "an", and "the" as used herein may also include the plural forms. The terms "comprising", "including", "containing", and "having" are inclusive and thus specify the presence of stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring them to be performed in the particular order described or illustrated, unless an execution order is explicitly stated. It should also be understood that additional or alternative steps may be used.

[0190] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features claimed herein.

Claims

1. An automatic material generation method, characterized in that, The method includes: Obtaining a target material to be added and a target prompt for generating an image, where the target prompt includes target area description information of a target area, and the target area does not include key content in the to-be-generated image; Invoking a preset image generation model, and using the target prompt to obtain a reference image including the target area; Marking the target area according to a marking box generation strategy to obtain a target image including at least one marking box; Adding the target material into the marking box according to the marking box in the target image to obtain a target material.

2. The method according to claim 1, wherein The marking the target area according to the marking box generation strategy to obtain a target image including at least one marking box includes: Obtaining the association relationship between an area and area description information; Determining the target area on the target image according to the target area description information and the association relationship; Using the marking box to mark the target area to obtain a target image including at least one marking box.

3. The method according to claim 1, wherein When the area description information includes area description sub-information corresponding to each target material, the adding the target material into the marking box according to the marking box in the target image to obtain a target material includes: Obtaining the area information corresponding to each marking box; Determining the marking box corresponding to each target material according to the area description sub-information of each target material and the area information corresponding to each marking box; Adding each target material into the corresponding marking box to obtain a target material.

4. The method according to claim 1, wherein The method further includes: Displaying a target image including at least one marking box; Receiving modification information of the marking box in the target image from a user; Modifying the marking box in the target image according to the modification information to obtain a target image including the modified marking box.

5. The method according to claim 1, wherein The target prompt further includes content requirements, and the invoking a preset image generation model and using the target prompt to obtain a reference image including the target area includes: Obtaining the content requirements in the target prompt; Invoking an image generation model, and using the content requirements and the area description information to obtain a reference image that meets the content requirements and the area description information.

6. The method according to claim 1, wherein The method further includes: Determining a reference area on an initial image; Adding a description to the reference area to obtain reference area description information; Establishing an association relationship between an area and area description information according to the reference area and the reference area description information.

7. The method according to claim 1 or 4, characterized in that, The method further includes: Detecting whether the marking box meets a preset condition on the target image; When the marking box does not meet the preset condition on the target image, processing the target image based on a preset image processing strategy.

8. An automatic material generation device, characterized in that, The apparatus includes: An obtaining unit, configured to obtain a target material to be added and a target prompt for generating an image, where the target prompt includes target area description information of a target area, and the target area does not include key content in the to-be-generated image; An invoking unit, configured to invoke a preset image generation model and use the target prompt to obtain a reference image including the target area; A marking unit, configured to mark the target area according to a marking box generation strategy, so as to obtain a target picture including at least one marking box; An adding unit, configured to add the target material into the marking box according to the marking box in the target picture, so as to obtain a target material.

9. An automatic material generation device, characterized in that, Comprising: At least one communication interface; At least one bus connected to the at least one communication interface; at least one processor connected to the at least one bus; At least one memory connected to the at least one bus, wherein the processor is configured to: Obtain a target material to be added and a target prompt word for generating a picture, the target prompt word includes target area description information of the target area, and the target area does not include key content in the picture to be generated; Call a preset picture generation model, and use the target prompt word to obtain a reference picture including the target area; Mark the target area according to a marking box generation strategy, so as to obtain a target picture including at least one marking box; Add the target material into the marking box according to the marking box in the target picture, so as to obtain a target material.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the material automatic generation method according to any one of claims 1 to 7.

Images