House content generation method and device based on artificial intelligence, and readable storage medium

Abstract

The embodiment of the invention discloses a house image content generation method based on artificial intelligence. The method comprises the steps of obtaining a basic image, wherein the basic image is an image shot in a house; processing the basic image based on a pre-trained cue word acquisition model to obtain a basic cue word; inputting the basic cue word and a preset cue word into an image generator to obtain a target image which is a realistic image; generating matched text content based on the target image; and the matched text content and the target image are associated and published to a target platform. Based on the scheme, on one hand, through full-process automatic design, links from image acquisition to publishing are linked, manual intervention is remarkably reduced, the content generation efficiency is improved, and the method is suitable for large-scale content production scenes of decoration companies, self-media and the like; on the other hand, by means of cue word design, it is ensured that the generated target image is highly close to a real house scene, meanwhile, the image-text collaborative generation mechanism guarantees consistency of content themes, and credibility and transmissibility are enhanced.

Description

Technical Field

[0001] The present application belongs to the field of artificial intelligence technology, and in particular relates to a method, device and readable storage medium for generating house content based on artificial intelligence. Background Art

[0002] Generative AI technologies (such as the text-based image model) are currently widely used in interior design and home improvement visualization. Social media platforms are increasingly demanding high-quality, authentic, and compliant interior design content. Users hope to use AI to generate highly realistic images of unfinished home renovations and publish them to various platforms. However, existing technologies lack an automated, compliant, and controllable AI image generation process.

[0003] Existing technologies have significant flaws in copyright and scene adaptation. For one thing, the copyright ownership of images of unfinished homes is unclear, and the direct use of online materials carries high legal risks, especially in commercial settings, which can easily lead to copyright disputes. Furthermore, general AI image generation systems are not optimized for unfinished homes, and the generated images often exhibit problems such as irrational spatial structure (e.g., abnormal wall proportions, misplaced doors and windows), and distorted material textures (e.g., overly smooth cement walls), making it difficult to meet the professional realism requirements of the renovation industry.

[0004] Existing technologies also have shortcomings in terms of process efficiency and compliance. First, there is a lack of automated capabilities for converting structural information from house images into high-quality prompts, resulting in a generation process that relies on manual experience, poor content consistency, and low efficiency. Second, the entire process from image generation to content publishing requires manual intervention at multiple stages (such as copywriting and platform adaptation), making automation impossible and restricting scalable operations. Third, content security mechanisms are weak, making it impossible to automatically review the compliance of images and copy, making it difficult to effectively detect sensitive elements, increasing the risk of platform penalties or content removal. Summary of the Invention

[0005] The purpose of the present invention is to provide a house content generation method based on artificial intelligence, aiming to solve the problem that the existing process of outputting house graphic content is relatively complicated and requires too much manual participation.

[0006] A first aspect of an embodiment of the present application provides a method for generating house image content based on artificial intelligence, comprising:

[0007] Acquire a basic image, where the basic image is an image taken inside a house;

[0008] Processing the basic image based on a pre-trained prompt word acquisition model to obtain basic prompt words;

[0009] Inputting the basic prompt words and the preset prompt words into an image generator to obtain a target image, wherein the target image is a realistic image;

[0010] generating matching text content based on the target image;

[0011] The matching text content and the target image are associated and published to a target platform.

[0012] Based on the artificial intelligence house image content generation method provided in the first aspect of the embodiment of the present application, optionally,

[0013] The basic image is any one of an unfinished house image, a fully decorated house image or a design effect image.

[0014] Based on the artificial intelligence house image content generation method provided in the first aspect of the embodiment of the present application, optionally, before processing the basic image based on the pre-trained prompt word acquisition model, the method further includes:

[0015] The basic image is preprocessed.

[0016] Based on the artificial intelligence-based house image content generation method provided in the first aspect of the embodiment of the present application, optionally, the preset prompt words include:

[0017] One or more of material prompt words, detail prompt words, character prompt words and atmosphere prompt words.

[0018] Based on the artificial intelligence-based house image content generation method provided in the first aspect of the embodiment of the present application, optionally, generating matching text content based on the target image includes:

[0019] The target image is input into a pre-trained text generation model to obtain matching text content.

[0020] Based on the artificial intelligence house image content generation method provided in the first aspect of the embodiment of the present application, optionally, the pre-trained text generation model is trained using text corpus whose popularity exceeds a preset value.

[0021] Based on the artificial intelligence-based house image content generation method provided in the first aspect of the embodiment of the present application, optionally, before publishing the matching text content and the target image to the target platform, the method further includes:

[0022] The matching text content and the target image are reviewed.

[0023] A second aspect of an embodiment of the present application provides an artificial intelligence-based house image content generation device, comprising:

[0024] an acquisition unit, configured to acquire a basic image, wherein the basic image is an image captured inside the house;

[0025] a processing unit, configured to process the basic image based on a pre-trained prompt word acquisition model to obtain basic prompt words;

[0026] An input unit, configured to input the basic prompt words and the preset prompt words into an image generator to obtain a target image, wherein the target image is a realistic image;

[0027] A generating unit, configured to generate matching text content based on a target image;

[0028] A publishing unit is used to publish the matching text content and the target image in association to a target platform.

[0029] Based on the artificial intelligence-based house image content generation device provided in the second aspect of the embodiment of the present application, optionally,

[0030] The basic image is any one of an unfinished house image, a fully decorated house image or a design effect image.

[0031] Based on the artificial intelligence-based house image content generation device provided in the second aspect of the embodiment of the present application, optionally, the processing unit is further configured to:

[0032] The basic image is preprocessed.

[0033] Based on the artificial intelligence-based house image content generation device provided in the second aspect of the embodiment of the present application, optionally, the preset prompt words include:

[0034] One or more of material prompt words, detail prompt words, character prompt words and atmosphere prompt words.

[0035] Based on the artificial intelligence-based house image content generation device provided in the second aspect of the embodiment of the present application, optionally, the generation unit is specifically configured to:

[0036] The target image is input into a pre-trained text generation model to obtain matching text content.

[0037] Based on the artificial intelligence-based house image content generation device provided in the second aspect of the embodiment of the present application, optionally, the pre-trained text generation model is trained using text corpus whose popularity exceeds a preset value.

[0038] Based on the artificial intelligence-based house image content generation device provided in the second aspect of the embodiment of the present application, optionally, the publishing unit is further configured to:

[0039] The matching text content and the target image are reviewed.

[0040] A third aspect of the embodiments of the present application provides an artificial intelligence-based house image content generation device, comprising:

[0041] CPU, memory, input and output interfaces, wired or wireless network interfaces, and power supply;

[0042] The memory is a transient storage memory or a persistent storage memory;

[0043] The central processing unit is configured to communicate with the memory and execute instruction operations in the memory on the device to perform the method as described in any one of the first aspects of the embodiments of the present application.

[0044] A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, characterized in that it includes instructions, which, when executed on a computer, enable the computer to execute the method described in any one of the first aspects of the embodiments of the present application.

[0045] As can be seen from the above technical solutions, the embodiments of the present application have the following advantages: The embodiments of the present application provide an artificial intelligence-based method for generating house image content, comprising: obtaining a base image, the base image being an image taken inside a house; processing the base image based on a pre-trained prompt word acquisition model to obtain basic prompt words; inputting the basic prompt words and preset prompt words into an image generator to obtain a target image, the target image being a realistic image; generating matching text content based on the target image; and publishing the matching text content and the target image to a target platform. The full-process automation and intelligence of this solution significantly improves production efficiency. The pre-trained prompt word acquisition model automatically analyzes house image features to generate basic prompt words, combines the preset prompt word template to drive the image generator to efficiently produce realistic target images, and simultaneously completes the intelligent generation and platform publishing of matching text content, completely abandoning the inefficient mode of traditional manual step-by-step operation, compressing the content generation cycle from several hours to minutes, reducing labor costs, and is particularly suitable for large-scale scenarios such as decoration companies producing design cases in batches and self-media accounts with daily updates. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. A person of ordinary skill in the art can also derive other drawings based on the provided drawings without inventive effort. It should be understood that the drawings provided in this section are only used to better understand the present solution and do not constitute a limitation of the present application.

[0047] Figure 1A flowchart of an embodiment of a method for generating house image content based on artificial intelligence provided by this application;

[0048] Figure 2 A schematic diagram of the basic image provided for this application;

[0049] Figure 3 A schematic diagram of the realistic image provided by this application;

[0050] Figure 4 Another schematic diagram of the base image provided for this application;

[0051] Figure 5 Another schematic diagram of the realistic image provided by this application;

[0052] Figure 6 This is a structural diagram of an embodiment of the method for generating house image content based on artificial intelligence provided by this application;

[0053] Figure 7 This is another structural schematic diagram of an embodiment of the artificial intelligence-based house image content generation method provided in this application. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application are clearly and completely described below. Obviously, the described embodiments are only 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 should fall within the scope of protection of this application. At the same time, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0055] The terms "first," "second," "third," "fourth," and the like (if any) in the specification and claims of this application and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions, e.g., a process, method, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus.

[0056] Generative AI technologies (such as the text-based image model) are currently widely used in interior design and home improvement visualization. Social media platforms are increasingly demanding high-quality, authentic, and compliant interior design content. Users hope to use AI to generate highly realistic images of unfinished home renovations and publish them to various platforms. However, existing technologies lack an automated, compliant, and controllable AI image generation process.

[0057] Existing technologies have significant flaws in copyright and scene adaptation. For one thing, the copyright ownership of images of unfinished homes is unclear, and the direct use of online materials carries high legal risks, especially in commercial settings, which can easily lead to copyright disputes. Furthermore, general AI image generation systems are not optimized for unfinished homes, and the generated images often exhibit problems such as irrational spatial structure (e.g., abnormal wall proportions, misplaced doors and windows), and distorted material textures (e.g., overly smooth cement walls), making it difficult to meet the professional realism requirements of the renovation industry.

[0058] Existing technologies also have shortcomings in terms of process efficiency and compliance. First, there is a lack of automated capabilities for converting structural information from house images into high-quality prompts, resulting in a generation process that relies on manual experience, poor content consistency, and low efficiency. Second, the entire process from image generation to content publishing requires manual intervention at multiple stages (such as copywriting and platform adaptation), making automation impossible and restricting scalable operations. Third, content security mechanisms are weak, making it impossible to automatically review the compliance of images and copy, making it difficult to effectively detect sensitive elements, increasing the risk of platform penalties or content removal.

[0059] To solve the above problems, this application provides a method for generating house image content based on artificial intelligence, please refer to Figure 1 An embodiment of the artificial intelligence-based house image content generation method provided in this application includes: steps 101 to 105.

[0060] 101. Obtain a basic image, wherein the basic image is an image taken inside the house.

[0061] Specifically, base images are images taken inside a house. These can be obtained through a variety of methods, such as from legitimate image websites and real estate databases. However, be mindful of copyright issues and ensure that the images you use have legal usage rights. For example, some public real estate datasets may contain interior images of various house types, allowing you to select suitable base images from these datasets.

[0062] 102. Process the basic image based on a pre-trained prompt word acquisition model to obtain basic prompt words.

[0063] Specifically, deep learning-based convolutional neural network (CNN) models such as ResNet and VGG can be used, as these models perform well in image feature extraction. Pre-trained models designed specifically for image prompt word generation, such as CLIP (Contrastive Language-Image Pretraining), can also be used. They can learn the association between images and text, thereby better extracting meaningful prompt words from images. For example, if the image shows a living room with floor-to-ceiling windows and wooden floors, the model may generate basic prompt words such as "living room, floor-to-ceiling windows, wooden floors." There is no limit to the specific type and training method of the prompt word acquisition model, as long as the basic prompt words can be effectively extracted from the basic image.

[0064] 103. Input the basic prompt words and the preset prompt words into an image generator to obtain a target image, wherein the target image is a realistic image;

[0065] Specifically, preset prompt words are pre-set according to different needs and scenarios to further enrich and adjust the style and details of the generated image. Preset prompt words can be set by professional designers or users according to their own preferences or business needs.

[0066] Preset prompts enhance the realism of target images through scene feature enhancement, physical rule constraints, and visual detail simulation. Specific implementation methods are as follows:

[0067] By injecting professional feature descriptions of the house scene, the image generator is constrained to output content that conforms to real-world logic and avoids surreal elements, such as adding prompts such as "exposed PVC pipes, uninstalled switch panels", etc.

[0068] By simulating the optical and material physical properties of the real world, the visual credibility of the image is improved, such as adding prompts such as "indoor diffuse reflection coefficient 0.7, ground reflectivity 20%".

[0069] Enhance the "lifelikeness" and credibility of images by adding real-world details that are easily perceived by the human eye, such as adding prompts such as "construction chalk lines on the wall, cement stains, etc."

[0070] You can use open-source image generation models, such as Stable Diffusion and DALL-E. These models have powerful image generation capabilities and can generate high-quality images based on input prompts. You can also develop a customized image generator based on your needs and adjust the model architecture and parameters to make it more suitable for generating house images.

[0071] The basic prompt word and the preset prompt word are combined and input into the image generator. The image generator will use the semantic information of the prompt word and its own learning capabilities to generate the target image that meets the requirements. During the generation process, the image generator parameters such as the number of generation steps and the random seed can be adjusted to control the detail and diversity of the generated image.

[0072] 104. Generate matching text content based on the target image.

[0073] Specifically, the team developed a series of text generation rules and templates. Based on the characteristics and category of the target image, they selected appropriate templates and filled them with the corresponding content. For example, for a target image of a bedroom, they could use a template like "This is a bedroom filled with [style] atmosphere. [Furniture 1] is neatly arranged, and [Furniture 2] adds a warm atmosphere." The content would then be filled in based on the actual style and furniture in the image.

[0074] Using pre-trained natural language processing models such as GPT-3 and BERT, the target image's description (which can be extracted using image recognition technology) is fed into the model, allowing it to generate matching text. After automatically generating the text, appropriate human editing and adjustments are performed to ensure accuracy, professionalism, and readability. For example, for content involving professional renovation terminology or descriptions of specific styles, human editing can provide more precise expression.

[0075] 105. Publish the matching text content and the target image in association to a target platform.

[0076] Target platforms can include social media platforms (such as Xiaohongshu, Weibo, and Douyin), interior design websites, and real estate sales platforms. Choose the appropriate platform for publishing based on your business needs and target audience. For example, if you're sharing renovation case studies for young consumers, Xiaohongshu might be a good choice; if a real estate developer is promoting a property, a real estate sales platform would be more appropriate. Before actual publication, specific software or manual review and archiving of published content is also possible, though these are not specified here.

[0077] As can be seen from the above technical solutions, the embodiments of the present application have the following advantages: The embodiments of the present application provide an artificial intelligence-based method for generating house image content, comprising: obtaining a base image, the base image being an image taken inside a house; processing the base image based on a pre-trained prompt word acquisition model to obtain basic prompt words; inputting the basic prompt words and preset prompt words into an image generator to obtain a target image, the target image being a realistic image; generating matching text content based on the target image; and publishing the matching text content and the target image to a target platform. The full-process automation and intelligence of this solution significantly improves production efficiency. The pre-trained prompt word acquisition model automatically analyzes house image features to generate basic prompt words, combines the preset prompt word template to drive the image generator to efficiently produce realistic target images, and simultaneously completes the intelligent generation and platform publishing of matching text content, completely abandoning the inefficient mode of traditional manual step-by-step operation, compressing the content generation cycle from several hours to minutes, reducing labor costs, and is particularly suitable for large-scale scenarios such as decoration companies producing design cases in batches and self-media accounts with daily updates.

[0078] The above content describes the specific implementation of this solution. Optionally, the implementation of this solution may also include multiple selective implementation methods, which are described in detail below.

[0079] (1) The basic image is any one of an unfinished house image, a fully decorated house image or a design effect image.

[0080] That is, adjust the image of rough house, fine decoration house or design effect. Take rough house image as an example. For details, please refer to Figures 2 to 5 ,in Figure 2 , Figure 4 As the base image, Figure 3 , Figure 5 The images processed based on this solution are adjusted based on different categories of images, thereby improving the targeting and effect of the output content.

[0081] (2) Before processing the basic image based on the pre-trained prompt word acquisition model, the method further includes: pre-processing the basic image.

[0082] Specifically, preprocessing includes copyright removal (such as cropping, texture reconstruction, etc.), as well as watermark removal, to eliminate potential copyright disputes in the base image, ensure the compliance of subsequent generation processes, and retain the image's scene information to the greatest extent possible, laying the foundation for prompt word collection and image regeneration.

[0083] (3) The preset prompt words include: one or more of material prompt words, detail prompt words, character prompt words and atmosphere prompt words.

[0084] Specifically, in the actual implementation process, different scenario groups can be set, corresponding to different prompt words respectively, and specific details are not limited here.

[0085] (4) Generating the matching text content based on the target image and the base image includes:

[0086] Inputting the target image into a pre-trained text generation model to obtain the matching text content.

[0087] The pre-trained text generation model is trained using text corpora with a popularity exceeding a preset value.

[0088] The pre-trained text generation model is trained using high-popularity corpora, and the popularity threshold can be customized. For example, for enterprise content, "view count > 500,000 + favorite count > 100,000" can be set as the standard for high-quality data, and for the self-media scenario, it can be reduced to "like count > 5,000". Or, in the actual implementation process, the method of identifying popular tags (such as "#old house renovation#decoration pitfalls") can be used to connect to third-party public opinion platforms (such as Baidu Index, Newrank) or the target platform API (such as the popular tag interface of Xiaohongshu) to collect corresponding text corpora.

[0089] Furthermore, the generated matching text content also has a popularity attribute. The text corpora input into the model have all been screened by popularity (such as indicators like view count, like count, favorite count, etc. being higher than the industry average), ensuring that the high-frequency words and sentence structures in the training set highly match the current user preferences.

[0090] Example: If the topic popularity of "light French decoration" on a certain platform has recently increased, and the training data contains a large number of highly praised copywriting containing words such as "light French, milky white, arched door", the model will automatically learn the combination rules of such expressions, and automatically call high-frequency words and sentence patterns when generating content. For example, for an image of rough house renovation, it will generate copywriting that conforms to the hot trend, such as "Old and shabby small house makeover|After knocking down three walls, 80㎡ actually squeezes out three bedrooms and two living rooms! Attached demolition and renovation list#Decoration counterattack".

[0091] (5) Review the matching text content and the target image.

[0092] Specifically, call a third-party review system (such as NetEase Yidun) and the internal review mechanism of the home improvement platform to conduct content security detection on the image and the copywriting; identify sensitive elements (such as political symbols, exposed scenes, dangerous construction operations), copyright watermarks, false propaganda content (such as images with extreme words like "absolutely environmentally friendly", "zero formaldehyde", etc.); and content for attracting traffic through cheating (such as off-site diversion information like WeChat / phone numbers). If a close-up of the face of a construction worker without blurring appears in the target image, the third-party system can automatically identify and trigger a "privacy risk" warning; if the text content contains words such as "most preferential", "exclusive agent", etc., the system will mark it as "advertising violation".

[0093] The above content describes the artificial intelligence-based house image content generation method provided by this application. To support the implementation of the above embodiment, this application also provides an artificial intelligence-based house image content generation device, please refer to Figure 6 , an embodiment of the artificial intelligence-based house image content generation device of the present application includes:

[0094] An acquisition unit 601 is configured to acquire a basic image, where the basic image is an image captured inside a house.

[0095] The processing unit 602 is configured to process the basic image based on a pre-trained prompt word acquisition model to obtain basic prompt words;

[0096] An input unit 603 is configured to input the basic prompt words and the preset prompt words into an image generator to obtain a target image, wherein the target image is a realistic image;

[0097] A generating unit 604 is configured to generate matching text content based on the target image;

[0098] The publishing unit 605 is configured to publish the matching text content and the target image in association to a target platform.

[0099] Optionally, the basic image is any one of an unfinished house image, a fully decorated house image or a design effect image.

[0100] Optionally, the processing unit is further configured to:

[0101] The basic image is preprocessed.

[0102] Optionally, the preset prompt words include:

[0103] One or more of material prompt words, detail prompt words, character prompt words and atmosphere prompt words.

[0104] Optionally, the generating unit is specifically configured to:

[0105] The target image is input into a pre-trained text generation model to obtain matching text content.

[0106] Optionally, the pre-trained text generation model is trained using text corpus whose popularity exceeds a preset value.

[0107] Optionally, the publishing unit is further configured to:

[0108] The matching text content and the target image are reviewed.

[0109] In this embodiment, the processes performed by each unit in the device are the same as those described above. Figure 1 The method processes described in the corresponding embodiments are similar and will not be repeated here.

[0110] Figure 7 1 is a structural diagram of an artificial intelligence-based house image content generation device 700 provided in an embodiment of the present application. The artificial intelligence-based house image content generation device 700 may include one or more central processing units (CPU) 701 and a memory 705, in which one or more applications or data are stored.

[0111] In this embodiment, the specific functional module division in the central processing unit 701 can be the same as the above Figure 6 The functional module division method of each unit described in is similar and will not be repeated here.

[0112] Memory 705 can be volatile or persistent storage. The program stored in memory 705 can include one or more modules, each of which can include a series of instruction operations on the server. Furthermore, the central processing unit 701 can be configured to communicate with memory 705, and execute the series of instruction operations in memory 705 on the artificial intelligence-based house image content generation device 700.

[0113] The artificial intelligence-based house image content generation device 700 may further include one or more power supplies 702 , one or more wired or wireless network interfaces 703 , and one or more input and output interfaces 707 .

[0114] The CPU 701 can execute the aforementioned Figure 1 The operations performed by the artificial intelligence-based house image content generation method in the illustrated embodiment will not be described in detail here.

[0115] An embodiment of the present application also provides a computer storage medium for storing computer software instructions used for the above-mentioned house image content generation method, which includes a program designed for executing the artificial intelligence-based house image content generation method.

[0116] The artificial intelligence-based method for generating house image content may be the artificial intelligence-based method for generating house image content described in the aforementioned figures.

[0117] An embodiment of the present application also provides a computer program product, which includes computer software instructions, and the computer software instructions can be loaded by a processor to implement the process of the artificial intelligence-based house image content generation method of any one of the above figures.

[0118] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the equivalent transformation of circuits and the division of units are only a kind of logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0120] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions or improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for generating house image content based on artificial intelligence, characterized in that: include: Acquire a basic image, where the basic image is an image taken inside a house; Processing the basic image based on a pre-trained prompt word acquisition model to obtain basic prompt words; Inputting the basic prompt words and the preset prompt words into an image generator to obtain a target image, wherein the target image is a realistic image; generating matching text content based on the target image; The matching text content and the target image are associated and published to a target platform.

2. The artificial intelligence house image content generation method according to claim 1, characterized in that: The basic image is any one of an unfinished house image, a fully decorated house image or a design effect image.

3. The method for generating house image content based on artificial intelligence according to claim 1, characterized in that: Before processing the basic image based on the pre-trained prompt word acquisition model, the method further includes: The basic image is preprocessed.

4. The method for generating house image content based on artificial intelligence according to claim 1, characterized in that: The preset prompt words include: One or more of material prompt words, detail prompt words, character prompt words and atmosphere prompt words.

5. The method for generating house image content based on artificial intelligence according to claim 1, characterized in that: Generating matching text content based on the target image includes: The target image is input into a pre-trained text generation model to obtain matching text content.

6. The method for generating house image content based on artificial intelligence according to claim 5, characterized in that: The pre-trained text generation model is trained using text corpus whose popularity exceeds a preset value.

7. The method for generating house image content based on artificial intelligence according to claim 1, characterized in that: Before publishing the matching text content and the target image in association with each other on the target platform, the method further includes: The matching text content and the target image are reviewed.

8. A device for generating house image content based on artificial intelligence, characterized in that: include: an acquisition unit, configured to acquire a basic image, wherein the basic image is an image captured inside the house; a processing unit, configured to process the basic image based on a pre-trained prompt word acquisition model to obtain basic prompt words; An input unit, configured to input the basic prompt words and the preset prompt words into an image generator to obtain a target image, wherein the target image is a realistic image; A generating unit, configured to generate matching text content based on a target image; A publishing unit is used to publish the matching text content and the target image in association to a target platform.

9. A device for generating house image content based on artificial intelligence, characterized in that: include: CPU, memory, input and output interfaces, wired or wireless network interfaces, and power supply; The memory is a transient storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory on the device to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The method comprises instructions, which, when executed on a computer, cause the computer to perform the method according to any one of claims 1 to 7.

Images