Image model training method and device for household appliances, server, and computer-readable storage medium

Abstract

The present application relates to the field of image processing technology, and discloses an image model training method for household appliances, including: fine-tuning a stable diffusion SD model according to a household appliance design image and a prompt word to obtain a first number of primary household appliance category basic models; performing a high-quality image effect test on the first number of primary household appliance category basic models, and screening out the primary household appliance category basic models that meet the test conditions as the second number of secondary household appliance category basic models; fusing the second number of secondary household appliance category basic models to obtain a household appliance design model; wherein the first number is greater than or equal to the second number. This method is conducive to making the household appliance design model meet the design needs of the household appliance industry and household appliance companies, and improving the intelligence level of AI automated design tools. The present application also discloses an image model training device, a server, and a computer-readable storage medium for household appliances.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, for example, to an image model training method and device for household appliances, a server, and a computer-readable storage medium. Background Art

[0002] At present, with the advancement and development of AI (Artificial Intelligence), there is an urgent need to apply AI to the field of design.

[0003] The related technology discloses an AIGC (Artificial Intelligence Generated Content) painting generation algorithm based on SD (Stable Diffusion) and controllable generation. The prompt words and sketches are input into the SD model, and the design drawings that meet the user's needs are output to realize AI automated design.

[0004] In the process of implementing the embodiments of the present disclosure, it is found that there are at least the following problems in the related art:

[0005] AI automated design tools cannot meet the design needs of the home appliance industry and home appliance companies and are not intelligent enough.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical components or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.

[0008] The embodiments of the present disclosure provide an image model training method and device, a server, and a computer-readable storage medium for home appliances, so that AI automated design tools can meet the design needs of the home appliance industry and home appliance companies and improve the level of intelligence.

[0009] In some embodiments, the method includes: fine-tuning the SD model according to the home appliance design image and prompt words to obtain a first number of primary home appliance category basic models; performing a high-quality image effect test on the first number of primary home appliance category basic models, and screening out the primary home appliance category basic models that meet the test conditions as the second number of secondary home appliance category basic models; fusing the second number of secondary home appliance category basic models to obtain a home appliance design model; wherein the first number is greater than or equal to the second number.

[0010] In some embodiments, the device includes: a primary model generation module, configured to fine-tune the SD model according to the home appliance design image and prompt words to obtain a first number of primary home appliance category basic models; a secondary model generation module, configured to perform a high-quality image effect test on the first number of primary home appliance category basic models, and screen out the primary home appliance category basic models that meet the test conditions as the second number of secondary home appliance category basic models; a home appliance design model generation module, configured to fuse the second number of secondary home appliance category basic models to obtain a home appliance design model; wherein the first number is greater than or equal to the second number.

[0011] In some embodiments, the device includes a processor and a memory storing program instructions, and the processor is configured to execute the image model training method for home appliances when running the program instructions.

[0012] In some embodiments, the server includes: a server body; and the image model training device for home appliances is installed on the server body.

[0013] In some embodiments, the computer-readable storage medium stores program instructions, and when the program instructions are executed, the computer is used to execute the image model training method for household appliances.

[0014] The image model training method and device for household appliances, server, and computer-readable storage medium provided in the embodiments of the present disclosure can achieve the following technical effects:

[0015] First, the SD model is fine-tuned according to the home appliance design images and prompt words to obtain the first number of primary home appliance category basic models, and then the first number of primary home appliance category basic models are tested for high-quality image effects to screen out the second number of secondary home appliance category basic models, and the second number of secondary home appliance category basic models are integrated to obtain the home appliance design model. This is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0016] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] One or more embodiments are exemplarily described by corresponding drawings, which do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements, and the drawings do not constitute a scale limitation, and wherein:

[0018] Figure 1is a schematic diagram of an image model training method for household appliances provided by an embodiment of the present disclosure;

[0019] Figure 2 is a schematic diagram of another image model training method for household appliances provided by an embodiment of the present disclosure;

[0020] Figure 3 is a schematic diagram of another image model training method for household appliances provided by an embodiment of the present disclosure;

[0021] Figure 4 is a schematic diagram of another image model training method for household appliances provided by an embodiment of the present disclosure;

[0022] Figure 5 is a schematic diagram of an image model training device for household appliances provided by an embodiment of the present disclosure;

[0023] Figure 6 is a schematic diagram of another image model training device for household appliances provided by an embodiment of the present disclosure;

[0024] Figure 7 It is a schematic diagram of a server provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] In order to be able to understand the features and technical contents of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings. The attached drawings are for reference only and are not used to limit the embodiments of the present disclosure. In the following technical description, for the convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.

[0026] The terms "first", "second", etc. in the specification and claims of the embodiments of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so that the embodiments of the embodiments of the present disclosure described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions.

[0027] Unless otherwise stated, the term "plurality" means two or more.

[0028] In the embodiment of the present disclosure, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B indicates: A or B.

[0029] The term "and / or" is a description of the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.

[0030] The term "correspondence" may refer to an association relationship or a binding relationship. The correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

[0031] Combination Figure 1 As shown, the embodiment of the present disclosure provides an image model training method for household appliances, comprising:

[0032] S101, the server fine-tunes the SD model according to the home appliance design image and prompt words to obtain a first number of primary home appliance category basic models.

[0033] S102, the server performs a high-quality image effect test on a first number of primary home appliance category basic models, and selects primary home appliance category basic models that meet the test conditions as a second number of secondary home appliance category basic models.

[0034] S103, the server fuses the second number of secondary home appliance category basic models to obtain a home appliance design model.

[0035] The first number is greater than or equal to the second number.

[0036] By using the image model training method for home appliances provided by the embodiment of the present disclosure, the SD model is first fine-tuned according to the home appliance design image and prompt words to obtain a first number of primary home appliance category basic models, and then the first number of primary home appliance category basic models are subjected to a high-quality image effect test to screen out a second number of secondary home appliance category basic models, and the second number of secondary home appliance category basic models are integrated to obtain the home appliance design model. This is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0037] Optionally, the server fine-tunes the SD model according to the home appliance design images and prompt words to obtain a first number of primary home appliance category basic models, including: the server collects home appliance design images. The server obtains corresponding prompt words according to the features of the home appliance design images. The server groups the home appliance design images according to the prompt words to obtain a first number of home appliance design image groups. The server inputs the i-th group of home appliance design images into the i-th SD model for training to obtain the i-th primary home appliance category basic model. Wherein, i=1, 2, ..., N; N is the first number.

[0038] Specifically, the server obtains corresponding prompt words according to the features of the home appliance design image, including: the server automatically annotates the home appliance design image using the features of the home appliance design image to obtain the corresponding prompt words.

[0039] In this way, corresponding prompt words are obtained according to the collected home appliance design images, and then grouped according to the prompt words to obtain multiple home appliance design image groups, and these home appliance design image groups are respectively input into the SD model for training to obtain the same number of primary home appliance category basic models as the home appliance design image groups. This is conducive to more accurately fine-tuning the SD model according to the home appliance design images and prompt words to obtain the first number of primary home appliance category basic models.

[0040] Optionally, the prompt word includes a home appliance category. The server groups the home appliance design images according to the prompt word to obtain a first number of home appliance design image groups, including: the server groups the home appliance design images of the same home appliance category based on a grouping rule to obtain the first number of home appliance design image groups. The grouping rule includes grouping the home appliance design images of the same home appliance category into a third number of groups.

[0041] Specifically, the value of the third number may be 30.

[0042] In this way, the home appliance design images are grouped, input into each SD model respectively, and the SD model is fine-tuned, which is conducive to more accurately fine-tuning the SD model according to the home appliance design images and prompt words, and obtaining the first number of primary home appliance category basic models.

[0043] Optionally, the server performs a high-quality image effect test on the first number of primary home appliance category basic models, and selects the primary home appliance category basic models that meet the test conditions as the second number of secondary home appliance category basic models, including: the server determines the primary output image corresponding to each primary home appliance category basic model. The server performs an effect test on the primary output image, and selects the primary output image that meets the test conditions as the secondary output image. The server determines the primary home appliance category basic model corresponding to the secondary output image as the second number of secondary home appliance category basic models.

[0044] In this way, the primary output image is tested for its effect, and the primary home appliance category basic model corresponding to the primary output image that meets the test conditions is screened out. This is conducive to better obtaining a secondary home appliance category basic model with better quality, so that the model meets the design needs of the home appliance industry and home appliance companies, and improves the intelligence level of AI automated design tools.

[0045] Optionally, the prompt word includes a category of home appliances. The primary output image satisfies the test condition, including: the home appliance product detail index of the primary output image satisfies the first test condition; and the prompt word index of the prompt word corresponding to the primary output image satisfies the second test condition. The home appliance product detail index includes one or more of lighting, light and shadow, composition, and no cluttered structure. The prompt word index includes one or more of color, material, texture detail, model, style, component, and category.

[0046] This is helpful to more accurately determine whether the detail indicators and prompt word indicators of the home appliance products meet the test conditions, thereby more accurately determining whether the primary output image meets the test conditions.

[0047] Optionally, the home appliance product detail index of the primary output image meets the first test condition, including: the server weights the home appliance product detail index of the primary output images of the same home appliance category, and the server uses the primary output image with the highest weighted proportion of the home appliance product detail index as the primary output image that meets the first test condition.

[0048] This is conducive to more accurately obtaining the quality evaluation effect of the home appliance product detail indicators of the primary output image, and whether the home appliance product detail indicators meet the test conditions, thereby more accurately judging whether the primary output image meets the test conditions.

[0049] Optionally, the prompt word index of the prompt word corresponding to the primary output image satisfies the second test condition, including: the server weights the prompt word index of the prompt word corresponding to the primary output image. The server uses the primary output image corresponding to the prompt word with the highest weighted proportion of the prompt word index as the primary output image that satisfies the second test condition.

[0050] In this way, it is helpful to more accurately obtain the quality evaluation effect of the prompt word index of the primary output image, and whether the prompt word index meets the test conditions, thereby more accurately judging whether the primary output image meets the test conditions.

[0051] Optionally, after the server determines the primary output image corresponding to each primary home appliance category basic model, the method further includes: the server performs an effect test on the primary output image, and selects the primary home appliance category basic model corresponding to the primary output image that does not meet the test condition as the home appliance category basic model to be optimized. The server optimizes the home appliance design image group and / or prompt words corresponding to the home appliance category basic model to be optimized.

[0052] In this way, for the primary home appliance category basic model corresponding to the primary output image that does not meet the test conditions, the corresponding home appliance design image group and / or prompt words are optimized, which is conducive to improving the quality of the training set and thus improving the quality of model training.

[0053] Optionally, the server optimizes the home appliance design image group and / or prompt words corresponding to the basic model of the home appliance category to be optimized, including: when the home appliance design image group corresponding to the home appliance category to be optimized does not meet the first test condition, the server replaces it with a new home appliance design image group by increasing the number of images or image quality.

[0054] Optionally, the server optimizes the home appliance image group and / or prompt words corresponding to the basic model of the home appliance category to be optimized, including: when the prompt words corresponding to the home appliance category to be optimized do not meet the second test condition, the server removes and redefines the prompt words corresponding to the home appliance category to be optimized.

[0055] In this way, if the home appliance design image group does not meet the first test condition, the number of images or the image quality is increased to replace a new home appliance design image group. If the prompt word does not meet the second test condition, the prompt word is removed and redefined. This is conducive to improving the quality of the training set, thereby improving the quality of model training.

[0056] Optionally, the server fuses the second number of secondary home appliance category basic models to obtain a home appliance design model, including: the server takes the secondary output images output by the second number of secondary home appliance category basic models as output, and the home appliance design images and prompt words corresponding to the secondary output images as input, to train the home appliance design model.

[0057] In this way, the secondary output image output by the secondary home appliance category basic model that meets the test conditions is used as the output, and the home appliance design image and prompt words corresponding to the secondary output image are used as input to train the home appliance design model, which is conducive to improving the quality of the training set and thus improving the training quality of the home appliance design model.

[0058] Optionally, the primary home appliance category basic model and the secondary home appliance category basic model are both Lora (Lora Low-Rank Adaptation of Large Language Models) models.

[0059] In this way, both the basic model of the primary home appliance category and the basic model of the secondary home appliance category are Lora models, which is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0060] Combination Figure 2 As shown, the embodiment of the present disclosure provides another image model training method for household appliances, including:

[0061] S201, the server collects home appliance design images.

[0062] S202: The server obtains corresponding prompt words according to the features of the home appliance design image.

[0063] S203: The server groups the home appliance design images according to the prompt words to obtain a first number of home appliance design image groups.

[0064] S204, the server inputs the i-th group of home appliance design images into the i-th SD model for training to obtain the i-th primary home appliance category basic model.

[0065] S205, the server performs a high-quality image effect test on the first number of primary home appliance category basic models, and selects the primary home appliance category basic models that meet the test conditions as the second number of secondary home appliance category basic models.

[0066] S206, the server merges the second number of secondary home appliance category basic models to obtain a home appliance design model.

[0067] The first number is greater than or equal to the second number. i=1, 2, ..., N; N is the first number.

[0068] Using the image model training method for home appliances provided by the embodiment of the present disclosure, first obtain the corresponding prompt words based on the collected home appliance design images, then group them according to the prompt words to obtain multiple home appliance design image groups, and input these home appliance design image groups into the SD model for training to obtain the same number of primary home appliance category basic models as the home appliance design image groups. This is conducive to more accurately fine-tuning the SD model according to the home appliance design images and prompt words to obtain the first number of primary home appliance category basic models. Then, the first number of primary home appliance category basic models are tested for high-quality image effects to screen out the second number of secondary home appliance category basic models, and the second number of secondary home appliance category basic models are integrated to obtain the home appliance design model. This is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0069] Combination Figure 3 As shown, the embodiment of the present disclosure provides another image model training method for household appliances, including:

[0070] S301, the server fine-tunes the SD model according to the home appliance design image and prompt words to obtain a first number of primary home appliance category basic models.

[0071] S302: The server determines a primary output image corresponding to each primary home appliance category basic model.

[0072] S303: The server performs an effect test on the primary output images, and selects primary output images that meet the test conditions as secondary output images.

[0073] S304: The server determines primary home appliance category basic models corresponding to the secondary output image as a second number of secondary home appliance category basic models.

[0074] S305, the server integrates the second number of secondary home appliance category basic models to obtain a home appliance design model.

[0075] The first number is greater than or equal to the second number.

[0076] By using the image model training method for home appliances provided by the embodiment of the present disclosure, the SD model is first fine-tuned according to the home appliance design image and the prompt words to obtain the first number of primary home appliance category basic models, and then the effect test is performed on the primary output image to screen out the primary home appliance category basic models corresponding to the primary output image that meets the test conditions, which is conducive to better obtaining secondary home appliance category basic models with better quality, and the second number of secondary home appliance category basic models are integrated to obtain the home appliance design model. It is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0077] Combination Figure 5 As shown, the embodiment of the present disclosure provides another image model training method for household appliances, including:

[0078] S501, the server fine-tunes the stable diffusion SD model according to the home appliance design image and the prompt words to obtain a first number of primary home appliance category basic models.

[0079] S502: The server determines a primary output image corresponding to each primary home appliance category basic model.

[0080] S503: The server performs an effect test on the primary output images, and selects primary output images that meet the test conditions as secondary output images.

[0081] S504: The server determines primary home appliance category basic models corresponding to the secondary output image as a second number of secondary home appliance category basic models.

[0082] S505, the server merges the second number of secondary home appliance category basic models to obtain a home appliance design model.

[0083] S506, the server performs an effect test on the primary output image, and selects the primary home appliance category basic models corresponding to the primary output images that do not meet the test conditions as the home appliance category basic models to be optimized.

[0084] S507, the server optimizes the home appliance design image group and / or prompt words corresponding to the basic model of the home appliance category to be optimized.

[0085] The first number is greater than or equal to the second number.

[0086] The image model training method for home appliances provided by the embodiment of the present disclosure is adopted. The SD model is first fine-tuned according to the home appliance design image and prompt words to obtain the first number of primary home appliance category basic models, and then the effect test is performed on the primary output image to screen out the primary home appliance category basic models corresponding to the primary output image that meets the test conditions, which is conducive to better obtaining secondary home appliance category basic models with better quality, and the second number of secondary home appliance category basic models are integrated to obtain the home appliance design model. For the primary home appliance category basic models corresponding to the primary output images that do not meet the test conditions, the corresponding home appliance design image group and / or prompt words are optimized, which is conducive to improving the quality of the training set, thereby improving the quality of model training. It is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0087] Combination Figure 5 As shown, the embodiment of the present disclosure provides an image model training device 200 for home appliances, including a primary model generation module 501, a secondary model generation module 502 and a home appliance design model generation module 503. The primary model generation module 501 is configured to fine-tune the stable diffusion SD model according to the home appliance design image and the prompt words to obtain a first number of primary home appliance category basic models. The secondary model generation module 502 is configured to perform a high-quality image effect test on the first number of primary home appliance category basic models, and screen out the primary home appliance category basic models that meet the test conditions as the second number of secondary home appliance category basic models. The home appliance design model generation module 503 is configured to fuse the second number of secondary home appliance category basic models to obtain a home appliance design model. Wherein, the first number is greater than or equal to the second number.

[0088] By using the image model training device 200 for home appliances provided by the embodiment of the present disclosure, the stable diffusion SD model is first fine-tuned according to the home appliance design image and the prompt words to obtain a first number of primary home appliance category basic models, and then the first number of primary home appliance category basic models are subjected to a high-quality image effect test to screen out a second number of secondary home appliance category basic models, and the second number of secondary home appliance category basic models are integrated to obtain the home appliance design model. This is conducive to making the home appliance design model meet the design needs of the home appliance industry and home appliance companies, and improving the intelligence level of AI automated design tools.

[0089] Combination Figure 6As shown, the embodiment of the present disclosure provides an image model training device 300 for household appliances, including a processor (processor) 600 and a memory (memory) 601. Optionally, the device 300 may also include a communication interface (Communication Interface) 602 and a bus 603. Among them, the processor 600, the communication interface 602, and the memory 601 can communicate with each other through the bus 603. The communication interface 602 can be used for information transmission. The processor 600 can call the logic instructions in the memory 601 to execute the image model training method for household appliances of the above embodiment.

[0090] In addition, the logic instructions in the memory 601 described above can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0091] The memory 601 is a computer-readable storage medium that can be used to store software programs and computer executable programs, such as program instructions / modules corresponding to the method in the embodiment of the present disclosure. The processor 600 executes the function application and data processing by running the program instructions / modules stored in the memory 601, that is, the image model training method for home appliances in the above embodiment is implemented.

[0092] The memory 601 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 601 may include a high-speed random access memory and may also include a non-volatile memory.

[0093] Combination Figure 7 As shown, an embodiment of the present disclosure provides a server 100, including: a server body, and the above-mentioned image model training device 200 (300) for household appliances. The image model training device 200 (300) for household appliances is installed on the server body. The installation relationship described here is not limited to placement inside the server body, but also includes installation connections with other components of the server 100, including but not limited to physical connections, electrical connections or signal transmission connections. It can be understood by those skilled in the art that the image model training device 200 (300) for household appliances can be adapted to a feasible server body, thereby realizing other feasible embodiments.

[0094] An embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above-mentioned image model training method for household electrical appliances.

[0095] The technical solution of the embodiment of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiment of the present disclosure. The aforementioned storage medium may be a non-transient storage medium, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes.

[0096] The above description and the accompanying drawings fully illustrate the embodiments of the present disclosure so that those skilled in the art can practice them. Other embodiments may include structural, logical, electrical, process and other changes. The embodiments represent only possible changes. Unless explicitly required, separate components and functions are optional, and the order of operation may vary. The parts and features of some embodiments may be included in or replace the parts and features of other embodiments. Moreover, the words used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and the claims, unless the context clearly indicates, the singular forms of "a", "an" and "the" are intended to include plural forms as well. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of listings containing one or more associated ones. In addition, when used in the present application, the term "comprise" and its variants "comprises" and / or comprising refer to the presence of stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof. In the absence of further restrictions, the elements defined by the sentence "comprising a ..." do not exclude the presence of other identical elements in the process, method or device comprising the elements. In this article, each embodiment may focus on the differences from other embodiments, and the same and similar parts between the various embodiments may refer to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, then the relevant parts can refer to the description of the method part.

[0097] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software may depend on the specific application and design constraints of the technical solution. The technicians may use different methods for each specific application to implement the described functions, but such implementations should not be considered to exceed the scope of the embodiments of the present disclosure. The technicians may clearly understand that, for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above may refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0098] In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units can be only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to implement this embodiment. In addition, each functional unit in the embodiment of the present disclosure may be integrated in a processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit.

[0099] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A method for training an image model for household appliances, characterized in that: include: Collect images of home appliance designs; According to the features of the home appliance design image, corresponding prompt words are obtained; grouping the home appliance design images according to the prompt words to obtain a first number of home appliance design image groups; Input the i-th group of home appliance design images into the i-th SD model for training to obtain the i-th primary home appliance category basic model; Wherein, i=1, 2, ..., N; N is the first number; Performing a high-quality image effect test on the first number of primary home appliance category basic models, and selecting the primary home appliance category basic models that meet the test conditions as the second number of secondary home appliance category basic models; A second number of secondary home appliance category basic models are integrated to obtain a home appliance design model; The first number is greater than or equal to the second number.

2. The method according to claim 1, characterized in that The prompt words include home appliance category; The home appliance design images are grouped according to the prompt words to obtain a first number of home appliance design image groups, including: based on the grouping rule, the home appliance design images of the same home appliance category are grouped to obtain a first number of home appliance design image groups; The grouping rule includes grouping the home appliance design images of the same home appliance category into a group with a third quantity.

3. The method according to claim 1, characterized in that A high-quality image effect test is performed on the first number of primary home appliance category basic models, and primary home appliance category basic models that meet the test conditions are screened out as the second number of secondary home appliance category basic models, including: determining a primary output image corresponding to each primary home appliance category base model; Performing an effect test on the primary output image, and selecting the primary output image that meets the test conditions as the secondary output image; The primary home appliance category base models corresponding to the secondary output images are determined as the second number of secondary home appliance category base models.

4. The method according to claim 3, characterized in that The prompt words include the categories of home appliances; the primary output images meet the test conditions, including: The home appliance product detail index of the primary output image meets the first test condition; and, The prompt word index of the prompt word corresponding to the primary output image satisfies the second test condition; Among them, the detail indicators of home appliance products include one or more of lighting, light and shadow, composition, and clutter-free structure; the prompt word indicators include one or more of color, material, texture details, model, style, components, and category.

5. The method according to any one of claims 1 to 4, characterized in that: The second number of secondary home appliance category basic models are integrated to obtain the home appliance design model, including: The secondary output images output by the second number of secondary home appliance category basic models are used as output, and the home appliance design images and prompt words corresponding to the secondary output images are used as input to train the home appliance design model.

6. An image model training device for household appliances, characterized in that: include: a primary model generation module configured to collect home appliance design images; According to the features of the home appliance design image, corresponding prompt words are obtained; grouping the home appliance design images according to the prompt words to obtain a first number of home appliance design image groups; Input the i-th group of home appliance design images into the i-th SD model for training to obtain the i-th primary home appliance category basic model; wherein i=1, 2, ..., N; N is the first number; A secondary model generation module is configured to perform a high-quality image effect test on a first number of primary home appliance category basic models, and select primary home appliance category basic models that meet the test conditions as a second number of secondary home appliance category basic models; a home appliance design model generating module, configured to fuse a second number of secondary home appliance category basic models to obtain a home appliance design model; The first number is greater than or equal to the second number.

7. An image model training device for household appliances, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the image model training method for household appliances as described in any one of claims 1 to 5 when running the program instructions.

8. A server, characterized in that: include: Server body; The image model training device for household appliances as described in claim 6 or 7 is installed on the server body.

9. A computer-readable storage medium storing program instructions, characterized in that: When the program instructions are executed, the computer is used to execute the image model training method for household appliances as described in any one of claims 1 to 5.

Images