Artificial intelligence-based packaging design generation method, device and storage medium

Abstract

The present application relates to artificial intelligence technology, and discloses a packaging design generation method, device and storage medium based on artificial intelligence. It includes: obtaining design reference data of a product; based on a first intelligent model, generating box design data using the design reference data; based on a second intelligent model, generating decorative design data using the design reference data and box design data; feeding back the decorative design data to the first intelligent model so that the first intelligent model can regenerate the box design data; updating the box design data in the second intelligent model so that the second intelligent model can regenerate the decorative design data; generating a packaging design result based on the latest box design data and the latest decorative design data. The present application also discloses a control device. The present application aims to enable an artificial intelligence model to simulate the process of a packaging designer integrating the box design and decorative design thinking, so as to generate a more complete and better packaging design result.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based packaging design generation method, a control device, a computer device, and a computer-readable storage medium. Background Art

[0002] Packaging design refers to the process of designing the appearance, structure, material and decoration of the packaging container for the product after the product is produced. Packaging design must not only meet the functions of product protection, transportation and storage, but also take into account the needs of product promotion, sales and brand communication. The most common one that can comprehensively meet the above needs is the packaging box. The packaging design of the product packaging box includes two parts: the box body design and the decorative design based on the box body.

[0003] With the development of artificial intelligence technology, the use of artificial intelligence models to automatically generate packaging design results can improve the efficiency of packaging design generation. At present, in the process of generating packaging design results by artificial intelligence models, the model is generally used to generate the box design, and then the corresponding decoration design is generated based on the box design.

[0004] However, this will make it difficult to consider the impact of decorative design on box design during the intelligent box design process, and it will be difficult to obtain a more complete and excellent packaging design result. This is because when packaging designers manually design the box, they can apply some of the ideas of decorative design to the box design, so that the box design and decorative design can influence each other, thereby obtaining a more complete and excellent packaging design result. However, it is difficult for current artificial intelligence models to do this, that is, it is difficult to simulate the thinking and creative process of packaging designers when integrating box and decorative design.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of this application is to provide a packaging design generation method, control device, computer equipment and computer-readable storage medium based on artificial intelligence, aiming to enable the artificial intelligence model to simulate the process of packaging designers integrating box body and decoration design thinking to generate more complete and excellent packaging design results.

[0007] To achieve the above objectives, the present application provides a packaging design generation method based on artificial intelligence, comprising the following steps:

[0008] Acquire design reference data of the product, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, product introduction information, and packaging design requirements;

[0009] Based on the first intelligent model, generating box body design data using design reference data;

[0010] Generate decoration design data based on the second intelligent model using the design reference data and the box body design data;

[0011] Feeding back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data using the design reference data and the decoration design data;

[0012] Updating the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data;

[0013] Generate packaging design results based on the latest box design data and the latest decoration design data.

[0014] Optionally, after the step of updating the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data, the step further includes:

[0015] Detecting whether a first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to a first threshold;

[0016] If yes, the step of generating a packaging design result according to the latest box design data and the latest decoration design data is executed.

[0017] Optionally, after the step of detecting whether the first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to a first threshold, the step further includes:

[0018] If not, the process returns to the step of feeding back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data using the design reference data and the decoration design data.

[0019] Optionally, after the step of detecting whether the first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to a first threshold, the step further includes:

[0020] If not, then the relevant feature used by the first intelligent model when the box body design data was most recently generated is updated to the first relevant feature, and the first intelligent model is made to extract the second relevant feature from the latest decoration design data, so as to regenerate the box body design data using the first relevant feature and the second relevant feature;

[0021] Return to the step of updating the box design data in the second intelligent model so that the second intelligent model regenerates the decoration design data.

[0022] Optionally, a first weight corresponding to the first relevant feature is greater than a second weight corresponding to the second relevant feature;

[0023] Each time the first intelligent model regenerates the box design data using the first relevant feature and the second relevant feature, the second weight is reduced.

[0024] Optionally, the step of generating the decoration design data based on the second intelligent model by using the design reference data and the box body design data includes:

[0025] Based on the second intelligent model, the shape features of each box surface are extracted from the box body design data, and the decorative features of each box surface are extracted from the design reference data, and the decorative design data of each box surface is generated in sequence;

[0026] Among them, starting from the second box surface, the third weight of the decorative design data of the previous box surface is adjusted when generating the decorative design data of the current box surface according to the second similarity of the decorative features between the current box surface and the previous box surface; and the higher the second similarity, the lower the third weight.

[0027] Optionally, the packaging design generation method based on artificial intelligence also includes:

[0028] Starting from the second box surface, after the decoration design data of the current box surface is generated, the continuity loss function is used to measure the smoothness between the decoration design data of the current box surface and the decoration design data of the previous box surface;

[0029] Detecting whether the smoothness meets preset requirements;

[0030] If not, the decoration design data of the current box surface is optimized so that the smoothness meets the preset requirements.

[0031] To achieve the above object, the present application also provides a control device, comprising:

[0032] An input module, used to obtain design reference data of a product, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, product introduction information, and packaging design requirements;

[0033] A first generating module, configured to generate box body design data using design reference data based on the first intelligent model;

[0034] A second generating module, for generating decoration design data based on a second intelligent model using design reference data and box body design data;

[0035] A feedback module, used to feed back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data using the design reference data and the decoration design data;

[0036] The second generating module is further used to update the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data;

[0037] The output module is used to generate packaging design results according to the latest box design data and the latest decoration design data.

[0038] To achieve the above objectives, the present application also provides a computer device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the computer program is executed by the processor, the steps of the above-mentioned artificial intelligence-based packaging design generation method are implemented.

[0039] To achieve the above objectives, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned artificial intelligence-based packaging design generation method are implemented.

[0040] The artificial intelligence-based packaging design generation method, control device, computer equipment and computer-readable storage medium provided in the present application use the design reference data of the product based on the artificial intelligence model to generate preliminary box design data and decoration design data in sequence, and then feed back the decoration design data to the box design model to regenerate the box design data, so that the box design and decoration design processes can influence each other to simulate the process of the packaging designer integrating the box design and decoration design thinking. In this way, not only can the artificial intelligence model be used to improve the packaging design efficiency of the product, but also a more complete and excellent packaging design result can be obtained, thereby improving the quality of packaging design. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of the steps of a packaging design generation method based on artificial intelligence in one embodiment of the present application;

[0042] Figure 2 This is a schematic diagram of the steps of a packaging design generation method based on artificial intelligence in another embodiment of the present application;

[0043] Figure 3 This is a schematic diagram of the steps of a packaging design generation method based on artificial intelligence in another embodiment of the present application;

[0044] Figure 4 This is a schematic diagram of the steps of a packaging design generation method based on artificial intelligence in yet another embodiment of the present application;

[0045] Figure 5 This is a schematic diagram of a control device in an embodiment of the present application;

[0046] Figure 6 A schematic diagram of the internal structure of a computer device according to an embodiment of the present application.

[0047] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0048] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0049] In addition, if the descriptions of "first", "second", etc. are involved in this application, they are only used for descriptive purposes (such as for distinguishing the same or similar features), and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in this field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0050] Reference Figure 1 In one embodiment, the packaging design generation method based on artificial intelligence includes:

[0051] Step S10: obtaining design reference data of the product, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, product introduction information, and packaging design requirements;

[0052] Step S20: Generate box body design data using design reference data based on the first intelligent model;

[0053] Step S30: Generate decoration design data based on the second intelligent model using the design reference data and the box body design data;

[0054] Step S40, feeding back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data using the design reference data and the decoration design data;

[0055] Step S50, updating the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data;

[0056] Step S60: Generate a packaging design result according to the latest box design data and the latest decoration design data.

[0057] In this embodiment, the execution terminal of the embodiment may be a computer device, or may be other equipment or devices (such as a control device) that controls the computer device.

[0058] Optionally, an artificial intelligence model for generating a packaging design of a product packaging box is constructed and trained in advance based on artificial intelligence technology to be applied in the packaging design generation method based on artificial intelligence. The artificial intelligence model includes a first intelligent model and a second intelligent model, and the first intelligent model is responsible for generating the box body design data of the packaging box, and the second intelligent model is responsible for generating the decoration design data of the packaging box.

[0059] Optionally, the first intelligent model focuses on generating the box body design data of the packaging box, and will model and learn the box body structure, size, etc. to adapt to the needs of packaging boxes of different types and shapes; the second intelligent model focuses on generating the decoration design data of the packaging box, and can learn how to design the appearance, pattern, color, etc. of the packaging box based on the design reference data and box body design data.

[0060] Optionally, the first intelligent model and / or the second intelligent model can be constructed based on machine learning or deep learning models such as neural networks, generative adversarial networks, and attention mechanisms.

[0061] As described in step S10, the model input is responsible for obtaining design reference data of the product to be packaged, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, a product introduction information and a packaging design requirement.

[0062] Optionally, the product image can be a picture of the product's appearance, used to show the product's shape, color and other characteristics; the product introduction information can include a detailed description of the product's features, materials, functions, uses, target audience, etc.; the packaging box draft drawing can be a preliminary framework drawing of the packaging box drawn by the customer or designer; the packaging design requirements can be the specific design requirements and expectations put forward by the customer, such as brand logo, style preference, market positioning, etc.

[0063] The design reference data input by the model can be carried in the form of text data and image data.

[0064] As described in step S20, the first intelligent model is pre-trained and learns the first association relationship between the relevant features of the design reference data and the box body design data. Based on the first association relationship, the first intelligent model has the ability to generate box body design data based on the design reference data. This training enables the model to more accurately understand the relevant features in the design reference data and generate box body design data accordingly.

[0065] During the training process, the first intelligent model will learn the first association between the features of the design reference data and the box design data. For example, the modeling and learning of the box structure, size, material, etc., the model minimizes the error between the predicted value and the actual value by adjusting the internal parameters. When the first intelligent model reaches a satisfactory performance level, it can be deployed in the production environment for practical application.

[0066] Optionally, the first intelligent model pre-processes the design reference data to ensure the uniformity and availability of the data format, which may include scaling and cropping of image data and standardization of text data. Then, during feature extraction, appropriate feature extraction methods may be used for different types of design reference data. For image data, convolutional neural networks may be used to extract visual features; for text data, natural language processing techniques may be used to extract key information.

[0067] Based on the extracted features, through model reasoning, the first intelligent model generates box design data that meets the requirements of the design reference data according to the first association relationship learned in the training stage.

[0068] Based on the results of model reasoning, the box design data can be preliminarily generated, including information such as the box structure, size, material, etc. These data will be adjusted and generated according to the characteristics and requirements of the design reference data to ensure that the box design meets the design requirements.

[0069] As described in step S30, the second intelligent model is pre-trained and learns the second association relationship between the design reference data, the relevant features of the box body design data and the decoration design data. Based on the second association relationship, the second intelligent model has the ability to generate decoration design data based on the design reference data and the box body design data. When the model reaches the expected performance level, it is deployed in the production environment for practical application.

[0070] Optionally, the design reference data and the box body design data are input into the trained second intelligent model. The design reference data expresses the needs and requirements of the design, while the box body design data includes information such as the structure and size of the box body.

[0071] Optionally, the feature extraction method of the second intelligent model for the design reference data can refer to the feature extraction method of the first intelligent model, which will not be repeated here; for the feature extraction of the box body design data, the second intelligent model uses the feature extractor in the deep learning model, such as a convolutional neural network or a self-attention mechanism, to encode the box body design data to extract its feature information. These feature extractors can capture important information such as the structure, shape, and size of the box body design data.

[0072] In addition, during the pre-training phase of the second intelligent model, the model will learn the second association between the design reference data, the relevant features of the box design data, and the decoration design data, and will also extract the features corresponding to the box design data. These features are usually embedded in the parameters of the model and can be obtained through the middle layer or weight matrix of the model.

[0073] If the design reference data and the box design data are multimodal (for example, images and text), the second intelligent model will also adopt a multimodal feature fusion method to fuse features from different data modalities to obtain a more comprehensive and rich feature representation.

[0074] Through model reasoning, the second intelligent model generates decoration design data based on the relevant features of the extracted design reference data and box design data, using the second association relationship learned in pre-training and fine-tuning.

[0075] According to the results of model reasoning, the decoration design data (including the decoration design data corresponding to each box surface) can be preliminarily generated, including the decoration style, pattern, color and other information of the box body. These data will be adjusted and generated according to the characteristics and requirements of the design reference data and the box body design data to ensure that the decoration design meets the overall design requirements.

[0076] As described in step S40, the first intelligent model is also pre-trained and learns the third association relationship between the relevant features of the design reference data, the decorative design data and the box body design data. Based on the third association relationship, the first intelligent model has the ability to regenerate the box body design data based on the design reference data and the decorative design data.

[0077] After the second intelligent model preliminarily generates the decoration design data, it will feed back the preliminarily generated decoration design data to the first intelligent model. When the first intelligent model receives the decoration design data fed back by the second intelligent model, in addition to extracting corresponding features from the decoration design data, it will also re-extract corresponding features from the design parameter data.

[0078] For feature extraction of the decoration design data, the first intelligent model can use a feature extractor in a deep learning model, such as a convolutional neural network or a self-attention mechanism, to encode the decoration design data, thereby extracting its feature information. In addition, during the pre-training phase of the first intelligent model, the model will learn the third association between the design reference data, the relevant features of the decoration design data, and the box body design data, and will also extract features corresponding to the decoration design data. These features are usually embedded in the parameters of the first intelligent model and can be obtained through the intermediate layer or weight matrix of the model.

[0079] If the design reference data and the decoration design data are multimodal (for example, images and text), the first intelligent model will also adopt a multimodal feature fusion method to fuse features from different data modalities to obtain a more comprehensive and rich feature representation.

[0080] Through model reasoning, the first intelligent model generates box design data based on the relevant features of the extracted design reference data and decoration design data, using the third association relationship learned in pre-training and fine-tuning.

[0081] In this way, it can be ensured that the first intelligent model can fully consider the correlation between the design reference data and the decoration design data when regenerating the box design data, add the decoration design to the box design process, and influence the box design. When simulating the packaging designer to manually design the box, the expected decoration design can be applied to the box design, making the two designs influence and blend with each other. That is, by comprehensively considering the characteristics of the design parameter data and the decoration design data, the first intelligent model can more accurately understand and simulate the entire design process, realize intelligent box design generation, and thus obtain more complete and excellent box design data.

[0082] As described in step S50, after the first intelligent model regenerates the box design data, the regenerated box design data is output to the second intelligent model. When the second intelligent model receives the box design data re-output by the first intelligent model, it updates and replaces the previous box design data with the latest received box design data.

[0083] The second intelligent model re-extracts features related to the box body design data from the latest box body design data, and combines them with features related to the previously extracted design reference data. Then, the decoration design data is regenerated based on the second association relationship between the design reference data, the relevant features of the box body design data and the decoration design data.

[0084] By outputting the regenerated box design data to the second intelligent model and updating the model according to the latest data, a dynamic design process can be achieved. The second intelligent model can regenerate the decoration design data according to the second association relationship between the design reference data, the relevant features of the box design data and the decoration design data, thereby forming a closed-loop system to continuously optimize and improve the decoration design data.

[0085] Optionally, the process of steps S40-S50 can be repeatedly executed as needed, so that the process of box body design and decoration design can repeatedly influence each other and be continuously updated in a cycle, thereby making the packaging design process more complete and excellent. By alternately updating the box body design data and the decoration design data, the process of the designer integrating the box body and decoration design thinking can be simulated, making the process of model packaging design closer to the process of human thinking, and improving the fidelity and creativity of the packaging design generated by the artificial intelligence model.

[0086] As described in step S60, in the model output, the final packaging design result is generated based on the latest generated box design data and decoration design data.

[0087] Optionally, based on the box body design data and the correlation between the decorative design data of each box surface, the box body design data and the decorative design data of each box surface are integrated together to form a unified packaging design style, that is, the decorative design data of each box surface is drawn onto each box surface in the box body design data to obtain the packaging design result of the product packaging box.

[0088] Optionally, the packaging design results can be output in the form of 3D graphics. Through 3D graphics, the three-dimensional structure and appearance characteristics of the packaging design can be clearly displayed, allowing users to more intuitively understand the overall effect of the product packaging box. In addition, in the 3D graphics, users can rotate, scale and move the model through interactive operations to browse and review the packaging design results more deeply, so as to better understand the design concepts and details.

[0089] In one embodiment, the design reference data of the product is used based on an artificial intelligence model to generate preliminary box design data and decoration design data in sequence, and then the decoration design data is fed back to the box design model to regenerate the box design data, so that the box design and decoration design processes can influence each other, so as to simulate the process of the packaging designer integrating the box design and decoration design thinking. In this way, not only can the artificial intelligence model be used to improve the packaging design efficiency of the product, but also a more complete and excellent packaging design result can be obtained, thereby improving the quality of packaging design.

[0090] In one embodiment, based on the above embodiment, referring to Figure 2 After the step of updating the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data, the method further includes:

[0091] Step S51, detecting whether a first similarity between the latest box body design data and the previous version of box body design data is greater than or equal to a first threshold;

[0092] Step S52: If yes, then execute the step of generating a packaging design result according to the latest box design data and the latest decoration design data.

[0093] In this embodiment, after the second intelligent model regenerates the decoration design data, the terminal may first detect whether the first similarity between the latest box design data and the previous version of the box design data is greater than or equal to the first threshold.

[0094] Optionally, various distance or similarity measurement methods may be used to calculate the similarity between two box design data, such as Euclidean distance, cosine similarity, Mahalanobis distance, etc.

[0095] After calculating the first similarity between the latest box body design data and the previous version of the box body design data, it is then detected whether the first similarity is greater than or equal to a first threshold value. The first threshold value is used to measure whether the first similarity reaches an expected value, and its value can be set according to actual needs, such as a value range of 90%-100%.

[0096] If it is detected that the first similarity is greater than or equal to the first threshold, it indicates that the range of update and adjustment of the latest box design data is already small, or it is no longer necessary to update and adjust (correspondingly, it also indicates that the range of update and adjustment of the latest decoration design data is already small, or it is no longer necessary to update and adjust). Therefore, at this time, it can be determined to directly execute the step of generating the packaging design result according to the latest box design data and the latest decoration design data (i.e., step S60), so as to generate the packaging design result according to the latest box design data and the latest decoration design data for output.

[0097] In this way, by judging the similarity of the design data and deciding whether to stop updating and adjusting it, unnecessary updates and adjustments can be avoided, saving system resources; and the similarity measurement method can be used to more accurately evaluate the differences between design data, avoiding excessive updates or preventing the model from falling into a local optimal solution.

[0098] In one embodiment, based on the above embodiment, referring to Figure 3 After the step of detecting whether the first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to the first threshold, the method further includes:

[0099] Step S53: If not, return to the step of feeding back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box design data using the design reference data and the decoration design data.

[0100] In this embodiment, after calculating the first similarity between the latest box design data and the previous version of the box design data, if it is detected that the first similarity is less than the first threshold, it indicates that the newly generated box design data still has a range that can be updated and adjusted, or it is necessary to update and adjust it again (correspondingly, it also indicates that the newly generated decorative design data also has a range that can be updated and adjusted, or it is necessary to update and adjust it again). Therefore, at this time, the step of feeding back the decorative design data to the first intelligent model so that the first intelligent model regenerates the box design data using the design reference data and the decorative design data (i.e., step S40) can be returned to execute, so that the latest decorative design data is fed back to the first intelligent model again, so as to cyclically execute steps S40-S50, and the box design data and the decorative design data can be continuously improved and optimized by repeatedly cyclically executing steps S40-S50, until the data before and after the update are almost the same, and then the packaging design result is generated and output according to the latest box design data and the latest decorative design data, so as to obtain a more complete and excellent product packaging design result.

[0101] In one embodiment, based on the above embodiment, referring to Figure 4 After the step of detecting whether the first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to the first threshold, the method further includes:

[0102] Step S54: if not, then update the relevant features used by the first intelligent model when generating the box body design data for the latest time to the first relevant features, and make the first intelligent model extract the second relevant features from the latest decoration design data, so as to regenerate the box body design data using the first relevant features and the second relevant features;

[0103] Step S55, returning to the step of updating the box design data in the second intelligent model so that the second intelligent model regenerates the decoration design data.

[0104] In this embodiment, after calculating the first similarity between the latest box body design data and the previous version of the box body design data, if it is detected that the first similarity is less than the first threshold, the relevant features (including the relevant features of the design reference data and the relevant features of the previous version of the decoration design data) used by the first intelligent model when the box body design data was most recently generated can be updated as the first relevant features. At the same time, the relevant features are extracted from the latest decoration design data as the second relevant features.

[0105] Then the first intelligent model regenerates the box design data again according to the first relevant feature, the second relevant feature, and the third association relationship.

[0106] After the first intelligent model regenerates the box design data, the regenerated box design data will be output to the second intelligent model, so as to return to execute the step of updating the box design data in the second intelligent model again so that the second intelligent model can regenerate the decoration design data (i.e., step S50), thereby allowing the second intelligent model to regenerate the decoration design data that can adapt to the latest version of the box design data.

[0107] In this way, the box design data and decoration design data can be updated and adjusted based on the previous box design data, and more perfect and excellent product packaging design results can be continuously optimized. This intelligent process can improve design efficiency and quality.

[0108] In one embodiment, based on the above embodiment, a first weight corresponding to the first related feature is greater than a second weight corresponding to the second related feature;

[0109] Each time the first intelligent model regenerates the box design data using the first relevant feature and the second relevant feature, the second weight is reduced.

[0110] In this embodiment, in the process of the first intelligent model regenerating the box design data according to the first relevant feature and the second relevant feature, corresponding weights are respectively assigned to the first relevant feature and the second relevant feature; wherein the first weight assigned by the first intelligent model to the first relevant feature is greater than the second weight assigned to the second relevant feature (e.g., the first weight: the second weight is 0.6:0.4).

[0111] In this way, the first intelligent model can pay more attention to the influence of the first related feature when re-updating and adjusting the box design data based on the previous version of the box design data. On the basis of retaining most of the box design data of the previous version, the box design data is re-adjusted to ensure that when optimizing the box design, the consistency of the previous and subsequent designs is maintained as much as possible.

[0112] Optionally, each time the first intelligent model regenerates the box design data using the first relevant feature and the second relevant feature, the second weight will be gradually reduced once (e.g., reduced by 0.05 each time until the value of the second weight is less than or equal to 0.05). This dynamic adjustment strategy can ensure that in the process of iterating the box design data multiple times, the influence of the second relevant feature gradually weakens, while the influence of the first relevant feature continues to strengthen (i.e., the first weight can be increased accordingly while the second weight is reduced), ensuring that each subsequent update of the box design data can inherit and retain most of the design style of the previous version of the box design data, that is, to extend the creation based on the design style of the previous version, thereby maintaining the consistency and stability of the box design.

[0113] In one embodiment, based on the above embodiment, the step of generating the decoration design data based on the second intelligent model using the design reference data and the box body design data includes:

[0114] Based on the second intelligent model, the shape features of each box surface are extracted from the box body design data, and the decorative features of each box surface are extracted from the design reference data, and the decorative design data of each box surface is generated in sequence;

[0115] Among them, starting from the second box surface, the third weight of the decorative design data of the previous box surface is adjusted when generating the decorative design data of the current box surface according to the second similarity of the decorative features between the current box surface and the previous box surface; and the higher the second similarity, the lower the third weight.

[0116] In this embodiment, the second intelligent model uses the design reference data and the box body design data to generate the decoration design data, and extracts the shape features (i.e., shape feature vectors) of each box surface from the box body design data. These features may involve information such as the geometric shape and size of the box body, and are used to describe the basic structural properties of each box surface.

[0117] Among them, in the box design data, the geometric shape features can include the shape type of the box surface, such as rectangle, square, circle, etc., and the properties of the corners, such as side length, curvature of the corners, etc. These features can be extracted by analyzing the structural data of the box body to describe the overall shape characteristics of the box surface.

[0118] Among them, the dimensional features refer to the various dimensional parameters of the box surface, such as length, width, height, etc. Through the dimensional information in the box design data, the dimensional features of each box surface can be accurately extracted, which helps to determine the proportional relationship and appearance size of the box body.

[0119] At the same time, the second intelligent model extracts the decorative features (i.e., decorative feature vectors) of each box surface from the design reference data. These features can include design elements such as color, texture, and pattern, which are used to describe the decorative style and design concept of each box surface.

[0120] Among them, through the second intelligent model, the color characteristics of each box surface can be extracted from the design information. The model can focus on the distribution and matching of color information in the design reference data, thereby describing the color theme, tonality and matching method of the box surface.

[0121] Among them, the second intelligent model can capture the details and features of the texture on the box surface. The model can focus on the style, thickness, texture and other features of the texture on different box surfaces, and describe the texture style and overall visual effect of the box surface.

[0122] Among them, the second intelligent model can extract the pattern features of each box surface from the design reference data. The model can focus on the type, layout, repetitiveness and other features of the pattern to describe the design style and uniqueness of the box surface pattern.

[0123] Combining decorative features such as color, texture, and pattern, the decorative style and design concept of each box surface can be described. By focusing on and emphasizing key design elements, the model can understand and extract the unique design style of the box surface, which helps to convey the concept and emotion that the packaging box design wants to express.

[0124] Alternatively, the second intelligent model may generate the decorative features of the total box surface based on the design reference data according to the total box surface area and shape, and then divide the decorative features of the total box surface into decorative features corresponding to each box surface according to the layout of each box surface.

[0125] Optionally, after obtaining the shape features and decorative features corresponding to each box surface, the second intelligent model further generates the decorative design data corresponding to each box surface in sequence according to the shape features and decorative features corresponding to each box surface. And after generating the decorative design data corresponding to the first box surface based on the shape features and decorative features corresponding to the first box surface, starting from the second box surface, the second intelligent model can also adjust the third weight of the decorative design data of the previous box surface when generating the decorative design data of the current box surface according to the second similarity of the decorative features between the current box surface and the previous box surface.

[0126] Among them, a lower second similarity means that the current box surface needs to pay more attention to the decorative design of the previous box surface in order to maintain the unity of the decorative design styles between different box surfaces; and a higher second similarity means that the current box surface can pay less attention to the decorative design of the previous box surface and pay more attention to other design elements, so as to maintain the unity of the decorative design style of the box surface while maximizing the diversity of the decorative designs of each box surface.

[0127] Optionally, when generating the decorative design data of the current box surface, the decorative design data of the previous box surface is used as one of the generating elements, and the second similarity of the decorative features between the current box surface and the previous box surface is compared with the second threshold value; wherein, if the second similarity is lower than the second threshold value, in the process of generating the decorative design data of the current box surface, the third weight corresponding to the decorative design data of the previous box surface is increased; if the second similarity is higher than the second threshold value, in the process of generating the decorative design data of the current box surface, the third weight corresponding to the decorative design data of the previous box surface is reduced; the value of the second threshold value can be 70%~90%, optionally 70%.

[0128] Through this processing method, the second intelligent model can dynamically adjust the corresponding third weight according to the similarity of the decorative features between the box surfaces, so that the decorative design data of each box surface not only takes into account the design features of the previous box surface, but also makes reasonable trade-offs based on the degree of similarity, thereby ensuring the consistency of the overall design style and the unity of the design style, while maximizing the diversity of the decorative design of each box surface.

[0129] In one embodiment, based on the above embodiment, the packaging design generation method based on artificial intelligence further includes:

[0130] Generate decorative design data of the first box surface according to the shape characteristics and decorative characteristics of the first box surface;

[0131] Starting from the second box surface, the shape feature and the decoration feature of the current box surface are marked as the third related feature and the fourth related feature respectively, and the decoration design data of the previous box surface is marked as the fifth related feature;

[0132] The decoration design data of the current box surface is generated according to the third related feature, the fourth related feature and the fifth related feature.

[0133] In this embodiment, the second intelligent model generates decorative design data suitable for the first box surface according to the shape characteristics and decorative features of the first box surface. At this stage, the second intelligent model will focus on and generate decorative design suitable for the box surface according to the shape characteristics and decorative features of the box surface.

[0134] Starting from the second box surface, the second intelligent model will mark the shape features and decorative features of the current box surface as the third relevant features and the fourth relevant features, and mark the decorative design data of the previous box surface as the fifth relevant feature. Then, based on these three relevant features, the decorative design data of the current box surface is automatically generated. In this way, the second intelligent model can take into account the decorative design data of the previous box surface and maintain the consistency and coherence of the overall design.

[0135] In this way, through the transmission and marking of features between different box surfaces, the second intelligent model can more flexibly generate corresponding decorative design data based on the features of the front and back box surfaces, thereby ensuring that the design of each box surface is coordinated with the overall design style, ensuring the continuity of the overall packaging design, and unifying the design style of each box surface.

[0136] In one embodiment, based on the above embodiment, before the step of generating the decoration design data of the current box surface according to the third related feature, the fourth related feature and the fifth related feature, it also includes:

[0137] adjusting the third weight of the fifth related feature according to the second similarity of the decorative features between the current box surface and the previous box surface;

[0138] The higher the second similarity is, the smaller the third weight of the adjusted fifth related feature is.

[0139] In this embodiment, the third weight is the attention weight corresponding to the fifth related feature (ie, the decorative feature of the previous box surface).

[0140] The step of adjusting, starting from the second box surface, the third weight of the decoration design data of the previous box surface when generating the decoration design data of the current box surface according to the second similarity of the decoration features between the current box surface and the previous box surface comprises:

[0141] Starting from the second box surface, before generating the decorative design data of the current box surface, the second intelligent model will calculate the second similarity of the decorative features between the current box surface and the previous box surface. This second similarity can measure the consistency between the decorative features of the two.

[0142] Among them, various distance or similarity measurement methods can be used to calculate the similarity between two decoration design data, such as Euclidean distance, cosine similarity, Mahalanobis distance, etc.

[0143] Optionally, based on the calculated second similarity, the second intelligent model will adjust the third weight of the fifth related feature accordingly. For example, if the second similarity is high, it means that the decorative features of the current box surface alone are sufficient to maintain the unity of the decorative design style with the previous box surface. Therefore, the third weight of the fifth related feature can be reduced, and more attention can be paid to other design elements (such as increasing the weight of the third related feature accordingly) to maximize the richness or diversity of the decorative design while maintaining the unity of the decorative design style between the box surfaces; on the contrary, if the second similarity is low, it means that it is difficult to maintain the unity of the decorative design style with the previous box surface based on the decorative features of the current box surface alone, then it is necessary to increase the third weight of the fifth related feature, and consider more the decorative design data of the previous box surface to maintain the unity of the decorative design styles between different box surfaces.

[0144] Optionally, there is an inverse proportional relationship between the second similarity and the third weight. Based on the inverse proportional relationship, after the second similarity is known, the third weight can be calculated. For example, if the inverse proportional factor (i.e., the proportional factor of the inverse proportional function) is 0.003, then when the second similarity is 100%, the third weight is 0.003, and when the second similarity is 1%, the third weight is 0.3.

[0145] Optionally, when adjusting the third weight, if the third weight needs to be lowered, the fourth weight corresponding to the third related feature can be adaptively increased (the fifth weight corresponding to the fourth related feature can remain unchanged or increase); if the third weight needs to be increased, the fourth weight can be adaptively lowered (the fifth weight can remain unchanged or increase).

[0146] In this way, a dynamic balance of attention weights for different features can be achieved.

[0147] In one embodiment, based on the above embodiment, before the step of adjusting the third weight of the fifth related feature according to the second similarity of the decorative features between the current box surface and the previous box surface, the method further includes:

[0148] Extracting the richness of design requirements from the text data based on semantic understanding;

[0149] According to the richness, adjusting the value of the inverse proportional factor between the second similarity and the third weight;

[0150] The higher the richness, the smaller the value of the inverse proportional factor.

[0151] In this embodiment, when processing text data, the richness information required for the design can be extracted from it through semantic understanding technology. This can be achieved through natural language processing technology and text mining methods to help understand the intentions and requirements behind the design.

[0152] Optionally, the amount and diversity of information contained in the design requirements described by text data can be used as indicators to measure richness. The amount of information refers to the amount of detailed and specific information contained in the design requirements, while diversity refers to the degree of diversity of different aspects or elements covered by the design requirements. On this basis, the richness of the design requirements can be quantified by using the corresponding richness assessment model based on semantic understanding.

[0153] First, the text data required by the design is processed and analyzed through semantic understanding technology. Semantic understanding can help the system understand the meaning of text data and identify the concepts, relationships and semantic information therein.

[0154] Extract key information, terms, requirements, constraints, etc. from the design requirements text. This can include using natural language processing techniques such as entity recognition, keyword extraction, topic modeling, etc.

[0155] The amount of information is measured by counting the number of unique terms, keyword frequency, information gain and other indicators in the design requirements. At the same time, the number and coverage of different topics, fields or aspects involved in the design requirements are analyzed to evaluate diversity.

[0156] Based on the above information, an evaluation model can be established to convert the amount and diversity of information into quantifiable indicators.

[0157] Optionally, the value of the inverse proportional factor between the second similarity and the third weight can be dynamically adjusted according to the extracted richness. Generally, the higher the richness of the design requirements, the lower the requirement for the uniformity of the design styles of each box surface, and the more emphasis is placed on the performance of the diversity of the design styles. Therefore, a smaller inverse proportional factor can be selected at this time to further reduce the adjustment value of the third weight; the lower the richness of the design requirements, the higher the requirement for the uniformity of the design styles of each box surface, so a larger inverse proportional factor can be selected at this time to further increase the adjustment value of the third weight.

[0158] In this way, by dynamically adjusting the inverse proportional factor, it is possible to better adapt to the degree of attention paid to the unity of the design style under different richness requirements, thereby achieving better design effects.

[0159] For example, the richness is set to three levels: low, medium, and high; the inverse proportional factor corresponding to the low-level richness is 0.003, the inverse proportional factor corresponding to the medium-level richness is 0.002, and the inverse proportional factor corresponding to the high-level richness is 0.001.

[0160] In practical applications, by continuously optimizing and adjusting the numerical relationship between richness and inverse proportional factor, combined with feedback information of actual design requirements and effects, the system can gradually learn more suitable parameter settings, thereby improving the efficiency and accuracy of the design process.

[0161] In one embodiment, based on the above embodiment, the packaging design generation method based on artificial intelligence further includes:

[0162] Starting from the second box surface, after the decoration design data of the current box surface is generated, the continuity loss function is used to measure the smoothness between the decoration design data of the current box surface and the decoration design data of the previous box surface;

[0163] Detecting whether the smoothness meets preset requirements;

[0164] If not, the decoration design data of the current box surface is optimized so that the smoothness meets the preset requirements.

[0165] In this embodiment, starting from the second box surface, after generating the decoration design data of the current box surface, the second intelligent model will use the continuity loss function to measure the smoothness between the decoration design data of the current box surface and the decoration design data of the previous box surface. This continuity loss function can quantify the visual continuity and smooth transition of the decoration design. By comparing the decoration design data between the current box surface and the previous box surface, the model can evaluate the consistency and coherence between them.

[0166] First, the decoration design data is converted into a representation form suitable for calculating the continuity loss, such as converting the decoration design data into a tensor or vector form for numerical calculation. After being converted into a tensor or vector form, a feature extraction operation can be further performed for use in numerical calculation.

[0167] The continuity loss function is used to calculate the smoothness between the decoration design data of the current box surface and the decoration design data of the previous box surface. The continuity loss function can be a function of various forms for focusing on the consistency and smoothness between two continuous data sequences, such as mean square error or absolute error, etc., to measure the degree of difference between the two data sequences.

[0168] Optionally, the calculated smoothness is tested to determine whether the smoothness of the decorative design data between the current box surface and the previous box surface meets the preset requirements. For example, the smoothness is tested to determine whether it reaches the preset smoothness, if so, it is determined to meet the preset requirements, if not, it is determined not to meet the preset requirements; wherein the preset smoothness represents the expected level of smoothness or consistency of the data, which can be determined by relevant engineers according to actual conditions.

[0169] Optionally, when it is detected that the smoothness of the decoration design data between the current box surface and the previous box surface does not reach a preset smoothness, the decoration design data of the current box surface is optimized so that the smoothness meets a preset requirement.

[0170] The packaging design generation method based on artificial intelligence also includes:

[0171] Optimize the current box surface decoration design data in a preset manner;

[0172] The preset method includes at least one of the following:

[0173] Adjust the transition between colors or patterns;

[0174] Modify the density and position of decorative elements;

[0175] Redesigned the light and dark changes in the transition area:

[0176] Optimize edge treatment of decorative elements.

[0177] Optionally, the smoothness and visual effect of the decorative design data can be optimized by adjusting the transition between colors or patterns, such as gradient, blending or fusion. For example, the transition between colors or patterns between box surfaces can be changed from abrupt change to smooth gradient (this can be achieved by adjusting the transition position, color value and gradient type (such as linear gradient, radial gradient, etc.) between colors).

[0178] Optionally, the density and position of the decorative elements can be adjusted to make their distribution on the box surface more uniform and reasonable to improve smoothness and visual balance. For example, an adaptive layout algorithm is used to intelligently adjust the position and density of the decorative elements according to the shape and size of the box surface to achieve the best smoothness and visual effect.

[0179] It should be noted that decorative elements refer to elements that constitute decorative patterns.

[0180] Optionally, the light and dark changes in the transition area can be redesigned to make the decorative design visually smoother and more unified. For example, a color gradient can be applied to the transition area, and the transition can be smoother by appropriately adjusting the starting and ending points, color, and transparency of the gradient; the transition area can be softened, retouched, or blurred to make the light and dark transition smoother and eliminate hard edges and discontinuities.

[0181] Optionally, the edges of decorative elements are optimized, such as smoothing, gradient or blurring, to improve the transition and integration effect between decorative elements. For example, the edges of decorative elements are slightly softened to reduce the abruptness of sharp edges and make the transition more natural and smooth; the saturation or brightness of the color is gradually reduced at the edge of the decorative element to achieve a smooth transition with the background or adjacent elements; the shape and details of the edge of the decorative element are adjusted to better blend with the surrounding environment; the transparency of the edge of the decorative element is adjusted to make it fade into the background or adjacent elements to avoid overly obvious edge boundaries; ensure that the color of the edge of the decorative element is coordinated with the color of the adjacent area or background to avoid unnatural color transitions.

[0182] In one embodiment, by optimizing the current box surface decoration design data so that its smoothness meets preset requirements, it can be ensured that the final generated overall packaging design result can maintain consistency and beauty in visual effects, thereby improving the quality and effect of the design.

[0183] In one embodiment, based on the above embodiment, the packaging design generation method based on artificial intelligence further includes:

[0184] When training the second intelligent model, the continuity loss function is added to the total loss function, and during the training process, minimizing the total loss function is used as one of the training objectives;

[0185] Among them, the model parameters are updated based on the back propagation algorithm to reduce the total loss function.

[0186] In this embodiment, when training the second intelligent model, adding the continuity loss function to the total loss function of the second intelligent model can help improve the performance and training effect of the model.

[0187] The total loss function consists of two parts, one is the loss function of the second intelligent model, and the other is the continuity loss function. The total loss function c is defined as follows:

[0188] c=a+λ*b;

[0189] Among them, a is the second intelligent model loss function; b is the continuity loss function; λ is a hyperparameter that weighs the importance between a and b, and is used to adjust the relative importance of the two parts of the loss function in training.

[0190] During the training process, minimizing the total loss function is taken as one of the training objectives, and the model parameters are updated through the back propagation algorithm to reduce the total loss function. In each round of training, the model parameters are updated by calculating the total loss function, so that the model can be effectively trained and optimized in terms of the second intelligent model and continuity.

[0191] Optionally, based on the back propagation algorithm, the gradients of the second intelligent model loss function and the continuity loss function to the model parameters are calculated respectively. Using the calculated gradient information, the gradient descent method or other optimization algorithms can be used to update the model parameters to reduce the total loss function. The specific method for updating the parameters can be stochastic gradient descent, momentum optimization (Momentum), AdaGrad, Adam, etc. Repeat the above steps, calculate the gradient of the total loss function and update the model parameters in each round of training until the training stop condition is reached (such as the model reaches the maximum number of iterations or converges to a certain degree).

[0192] In practical applications, the implementation of the backpropagation algorithm can be simplified with the help of automatic differentiation functions and optimizers provided by deep learning frameworks (such as TensorFlow, PyTorch, etc.). These frameworks can help efficiently calculate gradients, update model parameters, and provide tuning tools and functions to accelerate the model training process.

[0193] By combining the continuity loss function with the second intelligent model loss function, the model can be better guided to learn a more continuous and smooth attention distribution during the training process, thereby improving the performance of the model on sequence data. Adjusting the hyperparameter λ can balance the relative importance of the two parts of the loss function, so that the model can better adapt to the specific task requirements, that is, it can better unify and maintain the decorative design style of each box surface, thereby improving the overall quality and aesthetics of the decorative design.

[0194] In addition, refer to Figure 5 In an embodiment of the present application, a control device Z10 is further provided, comprising:

[0195] An input module Z11 is used to obtain design reference data of a product, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, product introduction information, and packaging design requirements;

[0196] A first generating module Z12, used to generate box body design data using design reference data based on the first intelligent model;

[0197] A second generating module Z13, for generating decoration design data based on the second intelligent model using the design reference data and the box body design data;

[0198] Feedback module Z14, used for feeding back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data by using the design reference data and the decoration design data;

[0199] The second generating module Z13 is further used to update the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data;

[0200] The output module Z15 is used to generate packaging design results according to the latest box design data and the latest decoration design data.

[0201] Optionally, the control device Z10 may be a virtual control device (such as a virtual machine) or a physical device (such as a physical device other than a computer device that can execute the corresponding method).

[0202] In addition, a computer device is also provided in an embodiment of the present application. The internal architecture of the computer device can be as follows: Figure 6 As shown, it includes a processor, a memory, a communication interface and an input interface connected by a system bus. Among them, the processor is used to provide computing and control capabilities. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database is used to store data called by the computer program. The communication interface is used to communicate data with an external terminal. The input interface is used to receive a signal input by an external device. When the computer program is executed by the processor, a packaging design generation method based on artificial intelligence as described in the above embodiment is implemented.

[0203] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the present application scheme, and does not constitute a limitation on the computer device to which the present application scheme is applied. For example, in some optional embodiments, the computer device may also include an output interface (not shown in the figure), and the output interface is also connected to the system bus and is used to output corresponding signals to the peripheral device.

[0204] In addition, the present application also proposes a computer-readable storage medium, the computer-readable storage medium includes a computer program, and when the computer program is executed by a processor, the steps of the packaging design generation method based on artificial intelligence as described in the above embodiment are implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0205] To summarize, the artificial intelligence-based packaging design generation method, control device, computer equipment and computer-readable storage medium provided in the embodiments of the present application use the design reference data of the product based on the artificial intelligence model to generate preliminary box design data and decoration design data in sequence, and then feed back the decoration design data to the box design model to regenerate the box design data, so that the box design and decoration design processes can influence each other to simulate the process of the packaging designer integrating the box design and decoration design thinking. In this way, not only can the artificial intelligence model be used to improve the packaging design efficiency of the product, but also a more complete and excellent packaging design result can be obtained, thereby improving the quality of packaging design.

[0206] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0207] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0208] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A packaging design generation method based on artificial intelligence, characterized in that: include: Acquire design reference data of the product, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, product introduction information, and packaging design requirements; Based on the first intelligent model, generating box body design data using design reference data; Based on the second intelligent model, the shape features of each box surface are extracted from the box body design data, and the decorative features of each box surface are extracted from the design reference data, and the decorative design data of each box surface is sequentially generated; wherein, starting from the second box surface, according to the second similarity of the decorative features between the current box surface and the previous box surface, when generating the decorative design data of the current box surface, the third weight for the decorative design data of the previous box surface is adjusted; and the higher the second similarity, the lower the third weight; Feeding back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data using the design reference data and the decoration design data; Updating the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data; Detecting whether a first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to a first threshold; If yes, generate the packaging design result according to the latest box design data and the latest decoration design data; If not, then the relevant feature used by the first intelligent model when generating the box body design data for the latest time is updated to the first relevant feature, and the first intelligent model is enabled to extract the second relevant feature from the latest decoration design data so as to regenerate the box body design data using the first relevant feature and the second relevant feature; return to execute the step of updating the box body design data in the second intelligent model so as to enable the second intelligent model to regenerate the decoration design data; wherein, the first weight corresponding to the first relevant feature is greater than the second weight corresponding to the second relevant feature; and each time the first intelligent model regenerates the box body design data using the first relevant feature and the second relevant feature, the second weight is reduced once.

2. The packaging design generation method based on artificial intelligence according to claim 1, characterized in that: The packaging design generation method based on artificial intelligence also includes: Starting from the second box surface, after the decoration design data of the current box surface is generated, the continuity loss function is used to measure the smoothness between the decoration design data of the current box surface and the decoration design data of the previous box surface; Detecting whether the smoothness meets preset requirements; If not, the decoration design data of the current box surface is optimized so that the smoothness meets the preset requirements.

3. A control device, characterized in that: include: An input module, used to obtain design reference data of a product, wherein the design reference data includes at least one of a product picture, a packaging box draft drawing, product introduction information, and packaging design requirements; A first generating module, configured to generate box body design data using design reference data based on the first intelligent model; A second generation module is used to extract the shape features of each box surface from the box body design data and the decorative features of each box surface from the design reference data based on the second intelligent model, and sequentially generate decorative design data of each box surface; wherein, starting from the second box surface, according to the second similarity of the decorative features between the current box surface and the previous box surface, a third weight for the decorative design data of the previous box surface is adjusted when generating the decorative design data of the current box surface; and the higher the second similarity, the lower the third weight; A feedback module, used to feed back the decoration design data to the first intelligent model, so that the first intelligent model regenerates the box body design data using the design reference data and the decoration design data; The second generating module is further used to update the box body design data in the second intelligent model so that the second intelligent model regenerates the decoration design data; An output module, detecting whether a first similarity between the latest box body design data and the previous version of the box body design data is greater than or equal to a first threshold; if so, generating a packaging design result according to the latest box body design data and the latest decoration design data; If not, then the relevant feature used by the first intelligent model when generating the box body design data for the latest time is updated to the first relevant feature, and the first intelligent model is enabled to extract the second relevant feature from the latest decoration design data so as to regenerate the box body design data using the first relevant feature and the second relevant feature; return to execute the step of updating the box body design data in the second intelligent model so as to enable the second intelligent model to regenerate the decoration design data; wherein, the first weight corresponding to the first relevant feature is greater than the second weight corresponding to the second relevant feature; and each time the first intelligent model regenerates the box body design data using the first relevant feature and the second relevant feature, the second weight is reduced once.

4. A computer device, characterized in that: The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of the packaging design generation method based on artificial intelligence as described in claim 1 or 2 are implemented.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the packaging design generation method based on artificial intelligence as claimed in claim 1 or 2 are implemented.

Images