Cloud artificial intelligence interaction system for professional pan-entertainment art industry

Abstract

The invention provides a cloud artificial intelligence interaction system for a professional pan-entertainment art industry, which relates to the field of art industry and comprises a basic parameter setting unit, an intelligent prompt word unit, an artificial intelligence creation auxiliary unit, a region selection unit, an intelligent paintbrush and eraser unit, a multi-layer management unit and a cloud storage unit. The system integrates the functions of multi-layer management, intelligent paintbrushes and erasers, selection optimization and the like, simplifies tedious operations in traditional creation, and greatly improves the efficiency of art creators; by means of automatic generation, style conversion and other tools, intelligent creation is achieved, manual adjustment of creators is greatly reduced, and therefore secondary editing and creation capacity is achieved, and the requirements of professional creators are met.

Description

Technical Field

[0001] The present invention relates to the field of fine arts industry, and in particular to a cloud-based artificial intelligence interaction system for the professional pan-entertainment fine arts industry. Background Art

[0002] First, in the pan-entertainment industry, traditional image processing software has high hardware requirements, lacks collaborative functions, and does not have intelligent creation assistance such as automatic generation and style conversion, which makes drawing difficult and time-consuming; secondly, the existing artificial intelligence drawing platform cannot intelligently optimize complex layer structures and does not have the drawing operations of traditional painting software, which means that the drawing platform can only generate images but does not have secondary editing and creation capabilities, making it difficult to meet the needs of professional creators. Summary of the invention

[0003] In order to make up for the above shortcomings, the present invention provides a method that aims to improve the problems in the prior art that traditional image processing software does not have intelligent creation assistance such as automatic generation and style conversion, and the existing artificial intelligence drawing platform cannot intelligently optimize complex layer structures and does not have the drawing operations of traditional painting software.

[0004] The present invention is achieved in that:

[0005] The present invention provides a cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry, including a basic parameter setting unit, an intelligent prompt word unit, an artificial intelligence creation auxiliary unit, a selection unit, an intelligent brush and eraser unit, a multi-layer management unit and a cloud-based storage unit;

[0006] The system usage process is also included, and the system usage process includes the following steps:

[0007] S1. Start creating and complete the setting of basic parameter setting units;

[0008] S2, upload the picture, apply the intelligent prompt word unit to get picture A;

[0009] S3, applying the artificial intelligence creation auxiliary unit to the obtained picture A to obtain picture B;

[0010] S4, saving the image B to the cloud storage unit and the multi-layer management unit;

[0011] S5. When image B needs to be edited again, the selection unit or the smart brush and eraser unit is used to complete the editing, and image C is obtained, which is saved to the multi-layer management unit;

[0012] S5. Save image C to the cloud storage unit and the multi-layer management unit.

[0013] Preferably, the basic parameter setting unit includes three modules, namely, a basic model module, a data setting module and a module for setting the number of generated images.

[0014] Preferably, the intelligent prompt word unit includes a system prompt word module, a picture uploading module, a positive prompt word picture selection module and a reverse prompt word picture selection module;

[0015] The output end of the picture uploading module is connected to the positive prompt word picture selecting module or the reverse prompt word picture selecting module.

[0016] Preferably, the artificial intelligence creation auxiliary unit includes a sketch optimization module and a style switching module. The sketch optimization module is used to extract the sketch according to the original image uploaded by the user; and the style switching module is used to select different styles for the generated sketch to create the image.

[0017] Preferably, the area selection unit combines an artificial intelligence algorithm to optimize the edges of irregular areas to form a selected area map with complete edges.

[0018] Preferably, the smart brush and eraser unit comprises a smart brush module and a smart eraser module;

[0019] The smart brush and eraser unit also includes an artificial intelligence analysis module and a deep learning model drawing module. The artificial intelligence analysis module analyzes the features of the original image, and then uses the deep learning model drawing module to redraw the image, optimize details or change the style and background elements.

[0020] Preferably, the multi-layer management unit includes a switching module, a downloading module, a copying module and a deleting module.

[0021] Preferably, S3 also includes generating a parameter judgment module; when graph B meets the requirements, proceed to S4; when graph B does not meet the requirements, return to S2.

[0022] Preferably, the graph C prepared by the area selection unit in S5 is returned to the generation parameter judgment module in S3.

[0023] Preferably, when the image B in S5 does not need secondary editing, it is saved to the cloud storage unit through manual triggering.

[0024] The beneficial effects of the present invention are:

[0025] This system integrates functions such as multi-layer management, smart brushes and erasers, and selection optimization, which simplifies the tedious operations in traditional creation and greatly improves the efficiency of art creators. It realizes intelligent creation through tools such as automatic generation and style conversion, greatly reducing the manual adjustments of creators, thus having secondary editing and creation capabilities to meet the needs of professional creators. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 This is a flow chart of an intelligent prompt word unit in a cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry provided by an embodiment of the present invention;

[0028] Figure 2 It is a flow chart of a cloud-based artificial intelligence interaction system for the professional pan-entertainment art industry provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Example

[0031] Reference Figure 1 and Figure 2 , a cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry, characterized by including a basic parameter setting unit, an intelligent prompt word unit, an artificial intelligence creation assistance unit, a selection unit, an intelligent brush and eraser unit, a multi-layer management unit and a cloud storage unit.

[0032] The basic parameter setting unit includes three modules, namely, a basic model module, a data setting module and a raw image quantity setting module.

[0033] Furthermore, the base model module can use LoRA, which is a lightweight model adaptation method that greatly reduces the amount of training parameters and storage overhead by inserting a low-rank matrix update method in a specific layer of the model. In the style transfer network, the goal of LoRA is to inject specific style information into the base generation model without retraining the entire model. Insert the LoRA module in the key layers of the base model (such as the self-attention layer, convolution layer, or decoder part).

[0034] The data setting module includes setting the size, sampling method and quantity information. The raw image quantity setting module is used to set the raw image quantity.

[0035] The intelligent prompt word unit includes a system prompt word module, a picture upload module, a positive prompt word picture selection module and a reverse prompt word picture selection module; the output end of the picture upload module is connected to the positive prompt word picture selection module or the reverse prompt word picture selection module.

[0036] Reference Figure 1 The system has built-in a series of commonly used prompt words, and classified them into different categories to facilitate users to make effective choices. At the same time, the system implements the prompt word reverse function (reverse prompt word selection module). The platform effectively extracts the keyword information of the picture based on the pictures uploaded by the user and generates the corresponding prompt words. At the same time, the prompt words can be used as positive prompt words or reverse prompt words when painting, so that the pictures generated by the platform contain or avoid the prompt word content, greatly improving the user's creation efficiency.

[0037] Most painting platforms in the prior art do not provide basic sketches, or only provide a single style and a small number of sketches, and users cannot create based on sketches. The artificial intelligence creation assistance unit in this system includes a sketch optimization module and a style switching module. The sketch optimization module is used to extract the sketch according to the original image uploaded by the user; the style switching module is used to select different styles for the generated sketch to create the image. Furthermore, through the artificial intelligence model + intelligent prompt words, users can quickly generate a layer of content; through manual creation assistance, a prototype sketch is generated, and different styles are combined to achieve style switching of the original image; through the selection unit, the local content of the layer is optimized and reconstructed; and then the finished content is processed and created through the smart brush and eraser unit. The creative results are sent to the cloud storage unit after being saved, and can be shared with other creators in the team.

[0038] Existing drawing software supports selection tools, but for complex and irregular shaped selections, traditional software often has difficulty in accurately selecting, and the edge optimization is relatively rigid. The selection unit described in this system combines artificial intelligence algorithms to optimize the edges of irregular areas and form a selected area map with complete edges.

[0039] In some embodiments, artificial intelligence algorithms (AI) automatically identify objects and accurately select areas in image editing, perform intelligent redrawing or style conversion, optimize details, and improve editing efficiency and creative flexibility. Edge optimization of irregular areas is achieved, allowing users to generate accurate selections with one click, especially in the efficiency and accuracy of selection of complex images, far exceeding traditional software. This greatly improves the efficiency of designers when editing complex images. When using it, users can generate accurate selections with one click according to the shape of the object, realizing ideal intelligent cutouts.

[0040] In existing drawing software, brushes and erasers lack adaptive functions. During use, users need to manually adjust the brush strokes, and the brush smoothing effect is rough. The smart brush and eraser unit described in this system includes a smart brush module and a smart eraser module, and the smart brush and eraser unit uses artificial intelligence to analyze the characteristics of the original image (such as lines, colors, textures, etc.), and then uses a deep learning model to redraw the image, optimize details, or change elements such as style and background, and finally generate a new work of art. Make the brush have an adaptive adjustment function. Users do not need to manually adjust the brush strokes when drawing, and can achieve a better smoothing effect, which can facilitate fine painting.

[0041] Drawing software in the prior art does not support layer sorting and batch management operations (such as downloading, copying, and deleting). In this system, the multi-layer management unit includes a switching module, a downloading module, a copying module, and a deleting module. With reference to the bidirectional linked list data structure, a layer list function is developed. When the user selects a model and clicks to generate multiple images, the platform will add the images to the layer list according to the generation order; with the help of the layer list function, users can perform batch operations that are not supported by conventional software, thereby improving the efficiency of layer operations.

[0042] Finally, refer to Figure 2 The use process of this system includes the following steps:

[0043] S1. Start creating and complete the setting of basic parameter setting units;

[0044] S2, upload the picture, apply the intelligent prompt word unit to get picture A;

[0045] S3, applying the artificial intelligence creation auxiliary unit to the obtained picture A to obtain picture B;

[0046] In addition, S3 also includes a parameter determination module; if the graph B meets the requirements, proceed to S4; if the graph B does not meet the requirements, return to S2;

[0047] S4, saving the image B to the cloud storage unit and the multi-layer management unit;

[0048] S5. When image B needs to be edited again, the selection unit or the smart brush and eraser unit is used to complete the editing, and image C is obtained, which is saved to the multi-layer management unit;

[0049] In addition, the graph C prepared by the selection unit in S5 is returned to the generation parameter judgment module in S3;

[0050] Secondly, when the image B in S5 does not need secondary editing, it is saved to the cloud storage unit through manual triggering;

[0051] S5. Save image C to the cloud storage unit and the multi-layer management unit.

[0052] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry, characterized by: It includes basic parameter setting unit, intelligent prompt word unit, artificial intelligence creation assistance unit, selection unit, intelligent brush and eraser unit, multi-layer management unit and cloud storage unit; The system usage process is also included, and the system usage process includes the following steps: S1. Start creating and complete the setting of basic parameter setting units; S2, upload the picture, apply the intelligent prompt word unit to get picture A; S3, applying the artificial intelligence creation auxiliary unit to the obtained picture A to obtain picture B; S4, saving the image B to the cloud storage unit and the multi-layer management unit; S5. When image B needs to be edited again, the selection unit or the smart brush and eraser unit is used to complete the editing, and image C is obtained, which is saved to the multi-layer management unit; S5. Save image C to the cloud storage unit and the multi-layer management unit.

2. According to claim 1, a cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry is characterized in that: The basic parameter setting unit includes three modules, namely, a basic model module, a data setting module and a generating image quantity setting module.

3. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: The intelligent prompt word unit includes a system prompt word module, a picture uploading module, a positive prompt word picture selection module and a reverse prompt word picture selection module; The output end of the picture uploading module is connected to the positive prompt word picture selecting module or the reverse prompt word picture selecting module.

4. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: The artificial intelligence creation auxiliary unit includes a sketch optimization module and a style switching module. The sketch optimization module is used to extract the sketch according to the original image uploaded by the user; the style switching module is used to select different styles for the generated sketch to create the image.

5. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: The selection unit combines artificial intelligence algorithms to optimize the edges of irregular areas to form a selected area map with complete edges.

6. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: The smart brush and eraser unit includes a smart brush module and a smart eraser module. The smart brush and eraser unit also includes an artificial intelligence analysis module and a deep learning model drawing module. The artificial intelligence analysis module analyzes the features of the original image, and then uses the deep learning model drawing module to redraw the image, optimize details or change the style and background elements.

7. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: The multi-layer management unit includes a switching module, a downloading module, a copying module and a deleting module.

8. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: The S3 also includes a parameter determination module; when the graph B meets the requirements, the process continues to S4; when the graph B does not meet the requirements, the process returns to S2.

9. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 8, characterized in that: The graph C prepared by the selection unit in S5 is returned to the generation parameter judgment module in S3.

10. The cloud-based artificial intelligence interactive system for the professional pan-entertainment art industry according to claim 1 is characterized in that: When the image B in S5 does not need secondary editing, it is saved to the cloud storage unit through manual triggering.

Images