Large Model-Based Poster Generation Method, Device, and Product

Abstract

The present disclosure provides a method, apparatus, electronic device, storage medium, and program product for generating posters based on large models, which relates to the field of artificial intelligence technology, specifically to the fields of large models and deep learning technology, and can be applied to the scenario of image generation. The specific implementation solution is as follows: determining indication data representing the requirements for generating posters through text and / or images; generating a poster through a pre-trained large model according to the indication data, where the large model is used to represent the correspondence between the indication data and the poster. The present disclosure reduces the threshold for generating posters and improves the flexibility and simplicity of the poster generation process.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to the fields of large models and deep learning technology. In particular, it relates to a method, device, electronic device, storage medium, and computer program product for generating posters based on large models, which can be applied to image generation scenarios. Background Art

[0002] Posters are involved in various scenarios in multiple industries such as greeting card production, advertising and marketing, product promotion, cultural promotion, event promotion, film and television promotion, public welfare preaching, self-media content operation, job recruitment, short video covers, and reading note sharing. When users create posters, they usually need to first collect or draw pictures, then create copywriting, and finally complete the creation of posters through typesetting and beautification. Ordinary users face high thresholds and low efficiency in manually creating high-quality posters. Summary of the Invention

[0003] The present disclosure provides a method, device, electronic device, storage medium, and computer program product for generating posters based on large models.

[0004] According to a first aspect, there is provided a method for generating a poster based on a large model, including: determining indication data representing the requirements for generating a poster through text and / or images; generating a poster through a pre-trained large model according to the indication data, where the large model is used to represent the correspondence between the indication data and the poster.

[0005] According to a second aspect, there is provided a device for generating a poster based on a large model, including: a determination unit configured to determine indication data representing the requirements for generating a poster through text and / or images; a generation unit configured to generate a poster through a pre-trained large model according to the indication data, where the large model is used to represent the correspondence between the indication data and the poster.

[0006] According to a third aspect, there is provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any implementation manner of the first aspect.

[0007] According to a fourth aspect, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method described in any implementation manner of the first aspect.

[0008] According to a fifth aspect, there is provided a computer program product, including: a computer program that, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0009] According to the technology of the present disclosure, a method and apparatus for generating posters based on a large model are provided. A user only needs to send instruction data representing the requirements for generating a poster to the artificial intelligence large model, and the artificial intelligence large model can generate a poster that meets the requirements for generating a poster according to the instruction data, reducing the threshold for generating posters and improving the flexibility and simplicity of the poster generation process.

[0010] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. Description of the Drawings

[0011] The drawings are used to better understand the present solution and do not constitute a limitation to the present disclosure. Among them:

[0012] Figure 1 is an exemplary system architecture diagram to which an embodiment according to the present disclosure can be applied;

[0013] Figure 2 is a flowchart of an embodiment of the method for generating a poster based on a large model according to the present disclosure;

[0014] Figure 3 is a schematic diagram of an application scenario of the method for generating a poster based on a large model according to this embodiment;

[0015] Figure 4 is a flowchart of another embodiment of the method for generating a poster based on a large model according to the present disclosure;

[0016] Figure 5 is a flowchart of another embodiment of the method for generating a poster based on a large model according to the present disclosure;

[0017] Figure 6 is a flowchart of another embodiment of the method for generating a poster based on a large model according to the present disclosure;

[0018] Figure 7 is a structural diagram of an embodiment of the apparatus for generating a poster based on a large model according to the present disclosure;

[0019] Figure 8 is a schematic diagram of the structure of a computer system suitable for implementing the embodiments of the present disclosure. Detailed Embodiments

[0020] The exemplary embodiments of the present disclosure will be described below in conjunction with the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, descriptions of well-known functions and structures are omitted in the following description for clarity and conciseness.

[0021] In the technical solution of the present disclosure, the processing of collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information complies with the provisions of relevant laws and regulations and does not violate public order and good customs.

[0022] Figure 1 An exemplary architecture 100 is shown to which the large model-based poster generation method and apparatus of the present disclosure can be applied.

[0023] As Figure 1 shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The terminal devices 101, 102, 103 are communicatively connected to form a topology network, and the network 104 is used as a medium to provide a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0024] The terminal devices 101, 102, 103 may be hardware devices or software that support network connections for data interaction and data processing. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices that support network connections, information acquisition, interaction, display, processing, etc., including but not limited to smartphones, tablets, e-book readers, laptop computers, and desktop computers, etc. When the terminal devices 101, 102, 103 are software, they may be installed in the above-listed electronic devices. It may be implemented as, for example, multiple software or software modules for providing distributed services, or it may be implemented as a single software or software module. No specific limitation is made here.

[0025] The server 105 may be a server that provides various services. For example, it is a background processing server that obtains the indication data representing the poster generation requirements sent by the terminal devices 101, 102, 103 and generates posters according to the indication data through an artificial intelligence large model. As an example, the server 105 may be a cloud server.

[0026] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers or as a single server. When the server is software, it can be implemented as multiple software or software modules (such as software or software modules for providing distributed services), or as a single software or software module. Specific limitations are not made here.

[0027] It should also be noted that the method for generating a poster based on a large model provided by the embodiments of the present disclosure can be executed by a server, or by a terminal device, or by the server and the terminal device cooperating with each other. Correspondingly, each part (such as each unit) included in the device for generating a poster based on a large model can be all arranged in the server, or all arranged in the terminal device, or respectively arranged in the server and the terminal device.

[0028] It should be understood that Figure 1 the numbers of the terminal devices, the network, and the server in

[0029] are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, the network, and the server. When the electronic device on which the method for generating a poster based on a large model runs does not need to perform data transmission with other electronic devices, the system architecture can only include the electronic device (such as a terminal device or a server) on which the method for generating a poster based on a large model runs.

[0029] Please refer to Figure 2 , Figure 2 which is a flowchart of a method for generating a poster based on a large model provided by the embodiments of the present disclosure. Among them, in process 200, the following steps are included:

[0030] Step 201, determine the indication data characterizing the poster generation requirements through text and / or images.

[0031] In this embodiment, the execution subject of the method for generating a poster based on a large model (such as Figure 1 the terminal device or the server in

[0032] The requirements for poster generation characterize the user's generation of a poster, which can be characterized by data in text form and / or image form. For example, the user can issue a poster generation requirement in the form of voice or text, and the above-mentioned execution entity recognizes the voice or text to obtain the poster generation requirement. Another example is that the user can hand-draw a simple version of the initial poster, and the above-mentioned execution entity determines the poster generation requirement by recognizing the initial poster. Another example is that the indication data includes text and images, and the user uses the two forms of data to more concretely characterize the poster generation requirement. In a specific example, the image is, for example, an architecture diagram of each element in the poster, and the text is, for example, requirement information characterizing the colors, styles, etc. of each element.

[0033] Posters can be of various scenes and types. For example, they are posters in scenarios such as greeting card making, advertising and marketing, product promotion, cultural promotion, event promotion, film and television promotion, public welfare preaching, self-media content operation, job recruitment, short video covers, reading note sharing, etc.

[0034] Step 202: Generate a poster based on the indication data through a pre-trained large model.

[0035] In this embodiment, the above-mentioned execution entity can generate a poster based on the indication data through a pre-trained large model. Among them, the large model is used to characterize the correspondence between the indication data and the poster.

[0036] The large model (artificial intelligence large model) specifically refers to an artificial intelligence pre-trained large model with a large-scale or ultra-large-scale number of parameters. The artificial intelligence large model includes two meanings. One is "pre-training", and the other is "large model". The combination of the two produces a new artificial intelligence mode, that is, after the model is pre-trained on a large-scale dataset, it can directly support various applications with little or no fine-tuning of a small amount of data. The artificial intelligence large model has excellent context understanding ability, language generation ability, learning ability, and transferability.

[0037] The above-mentioned execution entity can input the indication data into the artificial intelligence large model, extract features from the indication data through the artificial intelligence large model to obtain feature data; and then determine the semantic information characterized by the indication data in text form, the composition information, line information, etc. characterized by the indication data in image form based on the feature data, and generate a poster using the text-to-image and image-to-image capabilities.

[0038] In some alternative implementation manners of this embodiment, the indication data is the main title text characterizing the main title of the poster. The main title of the poster is the theme of the poster to be generated.

[0039] In this implementation manner, the above-mentioned execution entity can execute the above step 202 in the following manner:

[0040] First, through a large model, generate poster description text based on the main title text.

[0041] Poster description text is a written description used to introduce the content of a poster. It usually contains brief descriptions of elements such as the theme, images, and text in the poster, aiming to attract the attention of the audience and arouse their interest in the event, product, or service being promoted by the poster. For example, the content of the poster description text can include introductions to the time, location, participants, or special guests of the event, the featured functions of the product or service, as well as various eye-catching phrases or slogans. Its language should be concise, vivid, and appealing, capable of stimulating people's curiosity or emotional resonance, thus achieving the purpose of the poster.

[0042] In this implementation, the large model can understand the main title text based on its powerful natural language understanding ability, and then generate poster description text around the theme idea represented by the main title text.

[0043] Second, through a large model, generate a poster based on the main title text and the poster description text.

[0044] In this implementation, the large model determines information such as the style, layout, and color of the poster to be generated based on the main title text and the poster description text, and displays the main title text and the poster description text in the generated poster.

[0045] In this implementation, a specific implementation method for generating a poster based on the main title text is provided, which helps to further improve the flexibility and simplicity of the poster generation process.

[0046] In some alternative implementation methods of this embodiment, the poster description text includes a subtitle text and a copy introducing the poster content. In this implementation, the above-mentioned execution entity can execute the above first step in the following manner:

[0047] First, through a large model, generate the subtitle text based on the main title text; then, through a large model, generate the copy introducing the poster content based on the main title text and the subtitle text.

[0048] As an example, the large model can generate the subtitle text based on the main title text by referring to the generation methods in at least one of the following aspects:

[0049] Supplement the main title: Based on the main title text of the poster, think about further information or details related to the theme represented by the main title, and transform them into the subtitle. For example, if the main title is "Adventure Journey", the subtitle can be "The Brave Departure to Explore the Unknown Realm".

[0050] Emphasize uniqueness: Identify the most unique or prominent feature in the main title and highlight it in the subtitle. For example, if the main title is "The Wonders of Artistic Creativity", the subtitle could be "The Magic of Art Will Take You Beyond Imagination".

[0051] Convey emotions: Based on the emotion or mood conveyed by the main title, choose appropriate words to express and include them in the subtitle. For example, if the main title is "Blooming Hope", the subtitle could be "A Beautiful Journey to Feel the Power of Life".

[0052] Ask questions to guide: Transform the main title into an eye-catching question and provide an answer or solution in the subtitle. For example, if the main title is "How to Become a Successful Leader", the subtitle could be "Explore the Key Elements of Leadership and Discover Your Potential".

[0053] Create engaging language: Use attractive words, metaphors or vivid expressions to make the subtitle more vivid and interesting. For example, if the main title is "The Magic of Music", the subtitle could be "Notes Leaping, Hearts Resonating".

[0054] After determining the main title text and the subtitle text, the large model can combine the understanding of both to determine the key information, and then determine the key points to be emphasized in the introduction copy to generate the poster content introduction copy.

[0055] In this implementation, the above-mentioned execution entity generates the subtitle text based on the main title text, and then generates the poster content introduction copy based on both, so as to provide sufficient text data basis for the subsequent poster generation process, which helps to improve the accuracy and quality of the generated poster.

[0056] In some optional implementation manners of this embodiment, the above-mentioned execution entity can execute the generation process of the subtitle text in the following manner: First, through the large model, generate the initial subtitle text according to the main title text; then, adjust the initial subtitle text according to the received first editing operation to obtain the subtitle text.

[0057] The first editing operation can be issued by the user based on voice commands or action input commands. Taking the action input command as an example, each character in the generated initial subtitle text can be selected by the user based on operations such as clicking and touching, so that the user can conveniently determine the part to be modified in the initial subtitle text.

[0058] In this implementation, the user can manually adjust the initial subtitle text generated by the large model, which helps to further improve the matching degree of the subtitle text with the user's poster generation requirements and improve the quality of the poster to be generated.

[0059] In some alternative implementation manners of this embodiment, the above-mentioned execution entity may execute the generation process of the above-mentioned poster content introduction copy in the following manner: First, through a large model, according to the main title text and the subtitle text, generate an initial poster content introduction copy; then, according to the received second editing operation, adjust the initial poster content introduction copy to obtain the poster content introduction copy.

[0060] The second editing operation may be issued by the user based on voice instructions, action input instructions, etc. Taking the action input instruction as an example, each word in the generated initial poster content introduction copy can be selected by the user based on operations such as clicking and touching, so that the user can conveniently determine the part of the initial poster content introduction copy to be modified.

[0061] In this implementation manner, the user can manually adjust the initial poster content introduction copy generated by the large model, which helps to further improve the matching degree of the poster content introduction copy with the user's poster generation requirements and improve the quality of the poster to be generated.

[0062] In some alternative implementation manners of this embodiment, the above-mentioned execution entity may execute the above-mentioned second step in the following manner: Through a large model, according to the main title text, the subtitle text, and the poster content introduction copy, generate a poster.

[0063] As an example, the large model can determine information such as the style, layout, and color of the poster to be generated according to the main title text, the subtitle text, and the poster content introduction copy, so as to generate a poster, and display the main title text, the subtitle text, and the poster content introduction copy in the generated poster.

[0064] For the position information of the main title text, the subtitle text, and the poster content introduction copy in the poster, it can be determined through the poster template in the template library. As an example, the above-mentioned execution entity may determine the corresponding target template from the poster template library according to the main title text, the subtitle text, and the poster content introduction copy, so as to display the main title text, the subtitle text, and the poster content introduction copy in the poster according to the layout information of the target template.

[0065] In this implementation manner, the large model generates a poster based on more sufficient text data such as the main title text, the subtitle text, and the poster content introduction copy, further improving the accuracy and quality of the generated poster.

[0066] In some alternative implementation manners of this embodiment, the indication data is a draft poster diagram. In this implementation manner, the above-mentioned execution entity may execute the above-mentioned step 202 in the following manner:

[0067] First, through a large model, identify the draft poster diagram to obtain identification data.

[0068] As an example, the large model can extract features from the draft poster image to obtain a feature map; and then determine all the objects in the feature map, the positions of the objects, and the layout and interaction information between the objects, etc., that is, the information expressed by the draft poster image, to obtain recognition data. The interaction information between multiple objects includes but is not limited to occlusion relationships, collision relationships, dependency relationships, spatial relationships, temporal relationships, combination relationships, background relationships, etc.

[0069] Second, through the large model, generate a poster based on the recognition data.

[0070] As an example, the large model can respectively draw various objects included in the recognition data, and adjust and adapt each object according to the positions of the objects and the layout and interaction information between the objects, so that the objects are coordinated with each other in terms of color, style, etc., and adjust the shadow and light effects between adjacent objects with interactions to form a realistic effect.

[0071] In this implementation, an implementation method for a large model to generate a poster based on a draft poster image is provided. The user can hand-draw a draft poster image, and the large model can generate a high-quality poster, further reducing the design threshold of the poster and improving the flexibility and simplicity of the poster generation process.

[0072] In some optional implementation manners of this embodiment, the recognition data includes text data, line drawing data, and layout data. The text data is the text marked by the user for the whole or part in the draft poster image, such as main title text, subtitle text, poster content introduction copywriting, etc. text data, and the marked text such as advertising words on the objects in the draft poster image. The line drawing data is the line data in the draft poster image, and it is an important basis data for the large model to generate a poster based on the image-to-image generation ability.

[0073] In this implementation, the above execution subject can execute the above second step in the following manner: through the large model, generate a poster according to the text data, line drawing data, and layout data.

[0074] As an example, the large model uses the image-to-image generation ability to generate an initial poster according to the line drawing data and layout data; on the basis of the initial poster, add the text corresponding to the text data at the corresponding position to obtain the poster.

[0075] In this implementation, the large model generates a poster based on more comprehensive multi-type data such as text data, line drawing data, and layout data, further improving the accuracy and quality of the generated poster.

[0076] In some alternative implementation manners of this embodiment, the indication data includes a draft poster diagram and a first guiding text characterizing the adjustment direction of the draft poster diagram. The first guiding text may be an adjustment to aspects such as the style, structure, quantity, type, details, etc. of the objects in the draft poster diagram, or an adjustment to aspects such as the layout, interaction relationship, style, etc. of the overall draft poster diagram. For example, the first guiding text is "Use balance and layering to enhance the visual appeal of the poster."

[0077] In this implementation manner, the above-mentioned execution entity can execute the above step 202 in the following manner: Through a large model, generate a poster based on the draft poster diagram and the first guiding text.

[0078] As an example, the above-mentioned execution entity determines the adjustment direction indicated by the user according to the first guiding text through the large model, and generates a poster with reference to the adjustment direction during the poster generation process.

[0079] In this implementation manner, the user can simultaneously indicate to the large model the draft poster diagram and the first guiding text characterizing the adjustment direction of the draft poster diagram, and more accurately express the poster generation requirements to the large model in a combination of text and graphics, further improving the flexibility of the poster generation process and contributing to further improving the accuracy of poster generation.

[0080] In some alternative implementation manners of this embodiment, for the generated poster, the above-mentioned execution entity can also perform the following operations:

[0081] First, obtain a second guiding text for the modification direction of the poster.

[0082] The second guiding text may be an adjustment to aspects such as the style, structure, quantity, type, details, etc. of the objects in the generated poster, or an adjustment to aspects such as the layout, interaction relationship, style, etc. of the overall draft poster diagram.

[0083] Second, through the large model, modify the poster according to the second guiding text to obtain a modified poster.

[0084] In this implementation manner, the above-mentioned execution entity determines the modification direction indicated by the user according to the second guiding text through the large model, and on the basis of the generated poster, modifies the poster with reference to the adjustment direction to obtain a modified poster.

[0085] As an example, the second guiding text represents adding an object to the poster. Through an artificial intelligence large model, determine the specified object to be added and the adding position of the specified object in the object addition requirement; at the determined adding position, add the specified object; according to the image parsing result of the poster, adjust the specified object and the objects in the poster.

[0086] As an example, the large artificial intelligence model adjusts the specified object and the objects in the poster in the following ways:

[0087] According to the attribute information of the objects in the poster in the image parsing result, adjust the attribute information of the added specified object so that the attribute information of the added specified object is adapted to the attribute information of the objects in the poster.

[0088] According to the interaction relationship between the objects in the poster in the image parsing result, adjust the projection relationship, transition information, etc. of the interaction part between the added specified object and the objects in the poster.

[0089] In this implementation manner, a way for the user to modify the generated poster based on the second guiding text is provided, further improving the flexibility and convenience of the poster design process.

[0090] In some optional implementation manners of this embodiment, the above-mentioned execution subject may execute the above-mentioned first step in the following way:

[0091] First, determine the modification area targeted by the received selection operation from the poster; then, obtain the second guiding text for the modification direction of the modification area.

[0092] The selection operation may be represented based on the user's voice instruction, circle selection action instruction, etc.

[0093] In this implementation manner, the user can specifically modify a part of the area in the poster based on the second guiding text, further improving the flexibility and convenience of the poster design process.

[0094] In some optional implementation manners of this embodiment, the above-mentioned execution subject may also perform the following operations: First, obtain the editing operation for the poster; then, edit the poster according to the editing operation to obtain the edited poster.

[0095] Generally speaking, for users with the ability to design posters, they can send an editing operation to the above-mentioned execution subject to manually edit the poster. The editing operations include but are not limited to the modification of font, font size, text color, font bold, italic, underline, and the secondary adjustment of various attributes such as border, shadow, transparency, background color, alignment method, horizontal / vertical text arrangement, ordered / unordered numbering, line segment spacing, layer order, position, canvas size, etc.

[0096] In this implementation manner, providing the secondary editing ability for the generated poster helps to further improve the quality of the poster and meet the design requirements of professional users.

[0097] Continue to refer to Figure 3 , Figure 3FIG. 300 is a schematic diagram of an application scenario of the large model-based poster generation method according to this embodiment. In Figure 3 In the application scenario, the user 301 inputs indication data representing the poster generation requirements through text and / or images on the mobile terminal 302, and sends the indication data to the server 303. The server 303 first determines the indication data representing the poster generation requirements through text and / or images; then, through the pre-trained large model, according to the indication data, generates a poster. Among them, the large model is used to represent the correspondence between the indication data and the poster.

[0098] In this embodiment, a large model-based poster generation method and device are provided. The user only needs to send indication data representing the poster generation requirements to the artificial intelligence large model, and the artificial intelligence large model can generate a poster that meets the poster generation requirements according to the indication data, reducing the threshold of poster generation and improving the flexibility and simplicity of the poster generation process.

[0099] Continuing to refer to Figure 4 , FIG. 400 shows a schematic process of another embodiment of the large model-based poster generation method according to the present disclosure. In process 400, the following steps are included:

[0100] Step 401, determining the main title text representing the main title of the poster.

[0101] Step 402, generating a subtitle text through the pre-trained large model according to the main title text.

[0102] Step 403, generating a poster content introduction copy through the large model according to the main title text and the subtitle text.

[0103] Step 404, generating a poster through the large model according to the main title text, the subtitle text and the poster content introduction copy.

[0104] It can be seen from this embodiment that compared with the Figure 2 corresponding embodiment, the process 400 of the large model-based poster generation method in this embodiment specifically illustrates the process of the large model generating a poster based on the main title text input by the user, which helps to further improve the flexibility and convenience of the poster generation process and enhance the user experience in the image processing process.

[0105] Continuing to refer to Figure 5 , FIG. 500 shows a schematic process of another embodiment of the large model-based poster generation method according to the present disclosure. In process 500, the following steps are included:

[0106] Step 501, determining a draft poster diagram representing the poster generation requirements.

[0107] Step 502: Identify the draft poster image through the large model to obtain text data, line drawing data, and layout data.

[0108] Step 503: Generate a poster through the large model based on the text data, line drawing data, and layout data.

[0109] As can be seen from this embodiment, compared with Figure 2 the corresponding embodiment, the process 500 of the poster generation method based on the large model in this embodiment specifically illustrates the process of the large model generating a poster based on the draft poster image input by the user, which helps to further improve the flexibility and convenience of the poster generation process and enhance the user experience in the image processing process.

[0110] Continuing to refer to Figure 6 , a schematic process 600 of another embodiment of the poster generation method based on the large model according to the present disclosure is shown. In process 600, the following steps are included:

[0111] Step 601: Determine the draft poster image and the first guiding text characterizing the adjustment direction of the draft poster image.

[0112] Step 602: Generate a poster through the large model based on the draft poster image and the first guiding text.

[0113] As can be seen from this embodiment, compared with Figure 2 the corresponding embodiment, the process 600 of the poster generation method based on the large model in this embodiment specifically illustrates the process of the large model generating a poster based on the draft poster image and the guiding text input by the user, which helps to further improve the flexibility and convenience of the poster generation process and enhance the user experience in the image processing process.

[0114] Continuing to refer to Figure 7 , as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a poster generation device based on the large model. This device embodiment corresponds to Figure 2 the method embodiment shown, and this device can be specifically applied to various electronic devices.

[0115] As Figure 7 shown, the poster generation device 700 based on the large model includes: a determination unit 701 configured to determine the indication data characterizing the poster generation requirements through text and / or images; a generation unit 702 configured to generate a poster through a pre-trained large model based on the indication data, where the large model is used to characterize the correspondence between the indication data and the poster.

[0116] In some alternative implementation manners of this embodiment, the indication data is the main title text representing the main title of the poster, and the generating unit 702 is further configured to: generate a poster description text according to the main title text through a large model; and generate a poster according to the main title text and the poster description text through a large model.

[0117] In some alternative implementation manners of this embodiment, the poster description text includes a subtitle text and a poster content introduction copywriting, and the generating unit 702 is further configured to: generate a subtitle text according to the main title text through a large model; and generate a poster content introduction copywriting according to the main title text and the subtitle text through a large model.

[0118] In some alternative implementation manners of this embodiment, the generating unit 702 is further configured to: generate an initial subtitle text according to the main title text through a large model; and adjust the initial subtitle text according to the received first editing operation to obtain the subtitle text.

[0119] In some alternative implementation manners of this embodiment, the generating unit 702 is further configured to: generate an initial poster content introduction copywriting according to the main title text and the subtitle text through a large model; and adjust the initial poster content introduction copywriting according to the received second editing operation to obtain the poster content introduction copywriting.

[0120] In some alternative implementation manners of this embodiment, the generating unit 702 is further configured to: generate a poster according to the main title text, the subtitle text and the poster content introduction copywriting through a large model.

[0121] In some alternative implementation manners of this embodiment, the indication data is a poster draft diagram, and the generating unit 702 is further configured to: identify the poster draft diagram through a large model to obtain identification data; and generate a poster according to the identification data through a large model.

[0122] In some alternative implementation manners of this embodiment, the identification data includes text data, line drawing data and layout data, and the generating unit 702 is further configured to: generate a poster according to the text data, the line drawing data and the layout data through a large model.

[0123] In some alternative implementation manners of this embodiment, the indication data includes a poster draft diagram and a first guiding text representing the adjustment direction of the poster draft diagram; and the generating unit 702 is further configured to: generate a poster according to the poster draft diagram and the first guiding text through a large model.

[0124] In some alternative implementation manners of this embodiment, the above device further includes: a modification unit (not shown in the figure), configured to: obtain a second guiding text for the modification direction of the poster; modify the poster according to the second guiding text through a large model to obtain a modified poster.

[0125] In some alternative implementation manners of this embodiment, the modification unit is further configured to: determine, from the poster, a modification area targeted by the received selection operation; obtain a second guiding text for the modification direction of the modification area.

[0126] In some alternative implementation manners of this embodiment, the above device further includes: an editing unit (not shown in the figure), configured to: obtain an editing operation for the poster; edit the poster according to the editing operation to obtain an edited poster.

[0127] In this embodiment, a poster generation device based on a large model is provided. The user only needs to send indication data representing the poster generation requirement to the artificial intelligence large model, and the artificial intelligence large model can generate a poster that meets the poster generation requirement according to the indication data, reducing the poster generation threshold and improving the flexibility and simplicity of the poster generation process.

[0128] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor can implement the method for generating a poster based on a large model described in any of the above embodiments.

[0129] According to an embodiment of the present disclosure, the present disclosure further provides a readable storage medium, which stores computer instructions for enabling a computer to implement the method for generating a poster based on a large model described in any of the above embodiments when executed.

[0130] The embodiment of the present disclosure provides a computer program product, which can implement the method for generating a poster based on a large model described in any of the above embodiments when executed by a processor.

[0131] Figure 8FIG. 0 shows a schematic block diagram of an exemplary electronic device 800 that may be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as, for example, personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementations of the present disclosure described and / or claimed herein.

[0132] As Figure 8 shown, the device 800 includes a computing unit 801 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0133] A plurality of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0134] The computing unit 801 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 executes the various methods and processes described above, such as the large model-based poster generation method. For example, in some embodiments, the large model-based poster generation method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the large model-based poster generation method described above can be executed. Alternatively, in other embodiments, the computing unit 801 can be configured to execute the large model-based poster generation method by any other suitable means (e.g., by means of firmware).

[0135] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-a-chip systems (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0136] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0137] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0138] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0139] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0140] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of high management difficulty and weak business scalability existing in traditional physical hosts and virtual private server (VPS) services; it can also be a server of a distributed system, or a server combined with a blockchain.

[0141] According to the technical solution of the embodiment of the present disclosure, a method and an apparatus for generating a poster based on a large model are provided. A user only needs to send instruction data representing the poster generation requirement to the artificial intelligence large model, and the artificial intelligence large model can generate a poster that meets the poster generation requirement according to the instruction data, reducing the poster generation threshold and improving the flexibility and simplicity of the poster generation process.

[0142] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution provided by the present disclosure can be achieved, and no limitation is imposed herein.

[0143] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A poster generation method based on a large model, comprising: Determine instruction data required for poster generation represented by text and image, wherein the instruction data includes a poster draft image and a first guide text representing an adjustment direction of the poster draft image, and the first guide text is a guide text for an object in the poster draft image or a guide text for the entire poster draft image; Generate a poster according to the indication data through a pre-trained large model, including: using the large model to identify the poster draft image to obtain identification data, wherein the identification data includes line drawing data of the object, and layout data and interaction relationship data between the objects, and the line drawing data is line data in the hand-drawn poster draft image; using the large model to draw the object according to the line drawing data and the first guide text to obtain the drawn object, and adjust the drawn object according to the layout data and the interaction relationship data to generate the poster, wherein the large model is used to characterize the correspondence between the indication data and the poster.

2. The method according to claim 1, wherein: The indication data is the main title text representing the main title of the poster, and The method of generating a poster according to the indication data by using the pre-trained large model includes: Generate poster description text based on the main title text through the large model; The poster is generated through the large model according to the main title text and the poster description text.

3. The method according to claim 2, wherein: The poster description text includes the subtitle text and the poster content introduction copy, and The method of generating a poster description text according to the main title text through the large model includes: Generate the subtitle text according to the main title text through the large model; Through the large model, the poster content introduction copy is generated according to the main title text and the subtitle text.

4. The method according to claim 3, wherein: The step of generating the subtitle text according to the main title text through the large model includes: Generate an initial subtitle text based on the main title text through the large model; According to the received first editing operation, the initial subtitle text is adjusted to obtain the subtitle text.

5. The method according to claim 3, wherein: The method of generating the poster content introduction copy based on the main title text and the subtitle text through the large model includes: Generate an initial poster content introduction copy based on the main title text and the subtitle text through the large model; According to the received second editing operation, the initial poster content introduction copy is adjusted to obtain the poster content introduction copy.

6. The method according to claim 3, wherein: The step of generating the poster by using the large model according to the main title text and the poster description text includes: The poster is generated through the large model according to the main title text, the subtitle text and the poster content introduction copy.

7. The method according to claim 1, wherein: The identification data also includes text data, and The step of identifying the identification data through the large model and generating the poster includes: The poster is generated through the large model according to the text data, the line drawing data and the layout data.

8. The method according to any one of claims 1 to 7, wherein: Also includes: Obtaining a second guiding text for a modification direction of the poster; The poster is modified through the large model according to the second guide text to obtain a modified poster.

9. The method according to claim 8, wherein: The obtaining of a second guide text for a modification direction of the poster includes: Determining, from the poster, a modification area targeted by the received selection operation; The second guide text for the modification direction of the modification area is obtained.

10. The method according to any one of claims 1 to 7, wherein: Also includes: Obtaining editing operations for the poster; According to the editing operation, the poster is edited to obtain an edited poster.

11. A poster generation device based on a large model, comprising: A determination unit is configured to determine instruction data required for poster generation represented by text and image, wherein the instruction data includes a poster draft image and a first guide text representing an adjustment direction of the poster draft image, and the first guide text is a guide text for an object in the poster draft image or a guide text for the entire poster draft image; The generation unit is configured to generate a poster according to the indication data through a pre-trained large model, including: identifying the poster draft image through the large model to obtain identification data, wherein the identification data includes line drawing data of the object, and layout data and interaction relationship data between the objects, and the line drawing data is line data in the hand-drawn poster draft image; drawing the object through the large model according to the line drawing data and the first guide text to obtain a drawn object, and adjusting the drawn object according to the layout data and the interaction relationship data to generate the poster, wherein the large model is used to characterize the correspondence between the indication data and the poster.

12. The device according to claim 11, wherein The indication data is the main title text representing the main title of the poster, and The generating unit is further configured to: Through the large model, a poster description text is generated according to the main title text; through the large model, the poster is generated according to the main title text and the poster description text.

13. The device according to claim 12, wherein: The poster description text includes the subtitle text and the poster content introduction copy, and The generating unit is further configured to: Through the large model, the subtitle text is generated according to the main title text; through the large model, the poster content introduction copy is generated according to the main title text and the subtitle text.

14. The device according to claim 13, wherein: The generating unit is further configured to: The initial subtitle text is generated according to the main title text through the large model; and the initial subtitle text is adjusted according to the received first editing operation to obtain the subtitle text.

15. The device according to claim 13, wherein: The generating unit is further configured to: The initial poster content introduction copy is generated through the large model according to the main title text and the subtitle text; according to the received second editing operation, the initial poster content introduction copy is adjusted to obtain the poster content introduction copy.

16. The device according to claim 13, wherein: The generating unit is further configured to: The poster is generated through the large model according to the main title text, the subtitle text and the poster content introduction copy.

17. The device according to claim 11, wherein: The identification data also includes text data, and The generating unit is further configured to: The poster is generated through the large model according to the text data, the line drawing data and the layout data.

18. The device according to any one of claims 11 to 17, wherein: Also includes: A modification unit is configured to: obtain a second guide text for a modification direction of the poster; The poster is modified through the large model according to the second guide text to obtain a modified poster.

19. The device according to claim 18, wherein: The modification unit is further configured to: From the poster, determine the modification area targeted by the received selection operation; and obtain the second guide text for the modification direction of the modification area.

20. The device according to any one of claims 11 to 17, wherein: Also includes: The editing unit is configured to: obtain an editing operation for the poster; and edit the poster according to the editing operation to obtain an edited poster.

21. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

22. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 10.

23. A computer program product comprising: A computer program which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images