Marketing graphic and text generation system based on workflow construction

A workflow-based marketing graphic and text generation system addresses inefficiencies in manual production by using deep learning to quickly produce visually styled materials, enhancing user experience and accuracy.

TWI931730BActive Publication Date: 2026-07-11NAT CHIN YI UNIV TECH
0 Cites 0 Cited by

Patent Information

Application Number
TW113114821
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2024-04-19
Publication Date
2026-07-11
Estimated Expiration
2044-04-18

AI Technical Summary

Technical Problem

Existing marketing material production methods are time-consuming and inefficient, requiring manual operations and professional design skills, making it difficult for the general public to create visually styled materials.

Method used

A workflow-based marketing graphic and text generation system hosted on a server, utilizing a database, model building, and graphic/text generation modules, which includes deep learning to generate marketing materials based on user input and feedback, enabling quick and accurate production of materials in various styles.

Benefits of technology

The system efficiently generates marketing graphics and text materials that meet user needs, reducing time consumption and errors, and providing an intuitive interface for easy use by non-professionals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IMG-2_DRAW_113114821-A0101-14-0001-1
    Figure IMG-2_DRAW_113114821-A0101-14-0001-1
  • Figure IMG-2_DRAW_113114821-A0101-14-0002-2
    Figure IMG-2_DRAW_113114821-A0101-14-0002-2
  • Figure IMG-2_DRAW_113114821-A0101-14-0003-3
    Figure IMG-2_DRAW_113114821-A0101-14-0003-3
Patent Text Reader

Abstract

This application provides a marketing graphic and text generation system based on a workflow, comprising a database module, a model building module, and a graphic and text generation module that are coupled to each other. The database module stores multiple image data and multiple marketing text data with different visual styles; the model building module receives and performs deep learning on the image data and marketing text data to obtain a marketing graphic and text generation model; the graphic and text generation module has a startup unit, a generation unit, and an output unit based on the workflow. The startup unit allows the user to input a graphic and text requirement command, the generation unit generates multiple marketing graphic and text data of different styles according to the graphic and text requirement command using the marketing graphic and text generation model, and the output unit outputs the marketing graphic and text data for the user to select.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to a generation system, and more particularly to a marketing graphic generation system based on workflow construction. Prior Technology

[0002] The typical marketing materials (such as flyers, posters, etc.) produced by businesses usually involve using traditional drawing tools or manually operating drawing software to design the materials. They typically create rough drafts in various styles first, then select a final draft and use it to continue the subsequent production of the materials.

[0003] However, the existing methods of creating marketing materials still require a lot of manual operation, which makes them time-consuming and inefficient. Furthermore, designing images with a specific visual style requires professional design skills and software operation knowledge, making it difficult for the general public to create marketing materials on their own. Summary of the Invention

[0004] The main purpose of this application is to solve the problems of time-consuming and inefficient production methods in existing marketing materials.

[0005] To achieve the above objectives, one embodiment of this application provides a marketing graphic and text generation system based on a workflow. This system is hosted on a server, and a user connects to the server via a terminal device to use the system. The workflow-based marketing graphic and text generation system includes a database module, a model building module, and a graphic and text generation module. The database module stores multiple image data and multiple marketing text data with different visual styles. The model building module is coupled to the database module and receives and performs deep learning on the image data and marketing text data to obtain a marketing graphic and text generation model. The graphic and text generation module is coupled to the model building module and, based on the workflow, has a startup unit, a generation unit, and an output unit. The startup unit allows the user to input a graphic and text request command. The generation unit generates multiple marketing graphic and text data of different styles according to the graphic and text request command using the marketing graphic and text generation model. The output unit outputs the marketing graphic and text data for the user to select.

[0006] Therefore, the graphic and text generation module of this application can quickly generate marketing graphic and text materials corresponding to the user's needs according to the user's needs, thereby achieving the goal of solving the time-consuming and inefficient nature of existing marketing material production methods. Simple Explanation of the Diagram

[0007] [Figure 1] is a connection diagram of the marketing graphic generation system based on workflow construction according to an embodiment of this application. [Figure 2] is a block diagram of a marketing graphic generation system based on workflow construction according to an embodiment of this application. [Figure 3] is a block diagram of a marketing graphic generation system based on workflow construction according to another embodiment of this application. [Figure 4] is a schematic diagram of the implementation state of the interface module based on the workflow construction of this application embodiment, which is used to show that users can preview the marketing patterns in the marketing graphic materials. [Figure 5] is a schematic diagram of the implementation state of the interface module based on the workflow construction of this application embodiment, which is used to show that users can preview the marketing slogans in the marketing graphic materials. Implementation

[0008] To facilitate the explanation of the central idea expressed in the above-mentioned creative content section of this application, specific embodiments are provided below. The various objects in the embodiments are drawn to a scale suitable for illustrative purposes, rather than to the scale of actual components, as will be stated prior.

[0009] Please refer to Figures 1 to 5, which illustrate a marketing graphic generation system 100 based on a workflow structure according to an embodiment of this application. The system is hosted on a server 1. A user connects to the server 1 via a terminal device 200 to use the marketing graphic generation system 100. The workflow-based marketing graphic generation system 100 includes a database module 10, a model building module 20, and a graphic generation module 30. Furthermore, the marketing graphic generation system 100 includes an interface module 40, a feedback module 50, a personalized record module 60, and a communication module 70.

[0010] The database module 10 stores multiple image data and multiple marketing text data with different visual styles. In this embodiment, the image data includes images of existing known painting styles such as modern, retro, minimalist, sketch, and cartoon styles; the marketing text data includes multiple marketing slogans.

[0011] Please refer to Figures 1 to 3. In this embodiment of the application, the database module 10 has a storage unit 11. The storage unit 11 can store multiple marketing graphic data generated by the graphic generation module 30, feedback data from the feedback module 50, and selection record data recorded by the personalized record module 60, so that the subsequent model building module 20 can optimize the model based on the aforementioned data, and the user can also review and extract the aforementioned data.

[0012] A model building module 20, coupled to a database module 10, receives and performs deep learning on the image data and the marketing text data to obtain a marketing image and text generation model 21. In this embodiment, the deep learning method is a deep learning approach such as Generative Adversarial Networks (GAN) or Variational Autoencoder (VAE).

[0013] Please refer to Figures 1 to 3. In this embodiment of the application, the model building module 20 has an optimization unit 22. The optimization unit 22 can optimize the marketing graphic generation model 21 based on at least one of the feedback data, the selection record data, and the marketing graphic data, so that the marketing graphic data generated by the marketing graphic generation model 21 can better meet the needs of the user.

[0014] The graphic and text generation module 30, coupled to the model building module 20, has a startup unit 31, a generation unit 32, and an output unit 33 based on the workflow. The startup unit 31 allows the user to input a graphic and text request instruction; the generation unit 32 generates multiple marketing graphic and text materials of different styles using the marketing graphic and text generation model 21 corresponding to the graphic and text request instruction; the output unit 33 outputs the marketing graphic and text materials for the user to select. The marketing graphic and text materials may include marketing patterns or slogans of different styles.

[0015] In this embodiment, the graphic and text request instruction can be, for example, a text instruction such as "create a modern-style office chair promotional poster" or "generate a retro-style bicycle marketing flyer," for the subsequent generation unit 32 to perform corresponding generation actions. Furthermore, the user can also input further personalized instructions into the graphic and text request instruction, such as "create a modern-style office chair promotional poster and add a company logo" or "generate a cartoon-style bicycle marketing flyer suitable for elementary school students," thereby enabling the generation unit 32 to generate marketing graphic and text materials that better meet the user's needs.

[0016] Please refer to Figures 1 to 3. In this embodiment of the application, the graphic generation module 30 has a process control unit 34 based on the workflow. The process control unit 34 controls the start unit 31, the generation unit 32, and the output unit 33 to operate sequentially. Thus, through the process control unit 34, the graphic generation module 30 can quickly and accurately generate the marketing graphic materials that meet the user's needs according to the program.

[0017] Interface module 40 is coupled to graphic generation module 30. Referring to Figures 4 and 5, interface module 40 includes an input interface 41 and an output interface 42. Input interface 41 allows the user to access activation unit 31 and input the graphic request command; output interface 42 displays the marketing graphic data output by output unit 33. Thus, this application provides an intuitive and simple interface for the user through interface module 40, making it easier for the user to get started and obtain the marketing graphic data they need.

[0018] A feedback module 50, coupled to the model building module 20 and the graphic generation module 30, allows the user to provide feedback on the marketing graphic materials. In this embodiment, the feedback module 50 includes a rating unit 51 and a suggestion unit 52. The feedback includes a rating and a suggestion. The rating unit 51 allows the user to provide a rating on the marketing graphic materials; the suggestion unit 52 allows the user to provide suggestions on the marketing graphic materials. Thus, the feedback module 50 allows the user to provide feedback based on their usage experience, and outputs the feedback to the optimization unit 22 of the model building module 20. The optimization unit 22 can then optimize the marketing graphic generation model 21 based on the feedback, making the generated marketing graphic materials more suitable for the user's needs.

[0019] A personalized recording module 60, coupled to the model building module 20 and the graphic generation module 30, records selection data corresponding to the marketing graphic data selected by the user. The personalized recording module 60 records the marketing graphic data finally selected by the user and outputs the selection data to the optimization unit 22 of the model building module 20. The optimization unit 22 can then optimize the marketing graphic generation model 21 based on the selection data, making the generated marketing graphic data more suitable for the user's needs.

[0020] A communication module 70 is coupled to an interactive software 210 of the graphic generation module 30 and the terminal device 200. The user can access the startup unit 31 and input the graphic request command through the interactive software 210 and the communication module 70. The output unit 33 can output the marketing graphic data to the interactive software 210 through the communication module 70, and the interactive software 210 displays the marketing graphic data. The interactive software 210 can be any general communication software, but is not limited thereto; any software with interactive functions can be the interactive software 210.

[0021] Therefore, this application has the following advantages:

[0022] 1. The graphic and text generation module 30 of this application can quickly generate marketing graphic and text materials corresponding to the user's needs according to the user's needs, thereby achieving the purpose of solving the time-consuming and inefficient nature of existing marketing material production methods.

[0023] 2. This application enables the graphic generation module 30 to quickly and accurately generate marketing graphic materials that meet the user's needs according to the program through the process control unit 34, thereby reducing the probability of graphic generation errors in the graphic generation module 30.

[0024] 3. This application provides an intuitive and simple interface for the user through the interface module 40, making it easier for the user to get started and obtain the marketing graphic materials they need.

[0025] 4. The personalized record module 60 of this application can record the marketing graphic and text data finally selected by the user, and output the selection record data to the optimization unit 22 of the model building module 20. The optimization unit 22 can optimize the marketing graphic and text generation model 21 according to the selection record data, so that the generated marketing graphic and text data better meets the user's needs.

[0026] Although this application illustrates a preferred embodiment, those skilled in the art can make various modifications without departing from the spirit and scope of this invention. The embodiments described above are merely illustrative and not intended to limit the scope of this invention. All modifications or alterations made without violating the spirit of this invention fall within the scope of this patent application.

[0027] 1: Server 100: Marketing Graphics Generation System 200: Terminal device 210: Interactive Software 10: Database Module 11: Storage Unit 20: Model Building Module 21: Marketing Graphics Generation Model 22: Optimization Unit 30: Image and text generation module 31: Start-up Unit 32: Generation Unit 33: Output Unit 34: Process Control Unit 40: Interface Module 41: Input Interface 42: Output Interface 50: Feedback Module 51: Scoring Unit 52: Recommendation Unit 60: Personalized Recording Module 70: Communication Module

Claims

1. A marketing graphic and text generation system based on workflow construction, which is hosted on a server. A user connects to the server through a terminal device to use the marketing graphic and text generation system. The marketing graphic and text generation system based on workflow construction includes: a database module that stores multiple image data and multiple marketing text data with different visual styles; and a model building module coupled to the database module, which receives and performs deep learning on the image data and the marketing text data to obtain a marketing graphic and text generation model. A graphic and text generation module, coupled to the model building module, has a startup unit, a generation unit, and an output unit based on the workflow. The startup unit allows the user to input a graphic and text requirement instruction. The generation unit generates multiple marketing graphic and text materials of different styles according to the graphic and text requirement instruction using the marketing graphic and text generation model. The output unit outputs these marketing graphic and text materials for the user to select. A feedback module, coupled to the database module, the model building module, and the graphic and text generation module, allows the user to provide feedback on the marketing graphic and text materials.

2. The marketing graphic generation system based on workflow construction as described in claim 1, wherein, The image and text generation module has a process control unit based on the aforementioned workflow, which controls the start-up unit, the generation unit, and the output unit to operate sequentially.

3. The marketing graphic and text generation system based on workflow construction as described in claim 1 further includes an interface module coupled to the graphic and text generation module. The interface module includes an input interface and an output interface. The input interface is used to allow the user to enter the startup unit and input the graphic and text request command. The output interface is used to display the marketing graphic and text data of the output unit.

4. The marketing graphic generation system based on workflow construction as described in claim 1, wherein, The feedback module includes a rating unit and a suggestion unit. The feedback data includes rating data and suggestion data. The rating unit is used to allow the user to provide the rating data for the marketing graphic materials, and the suggestion unit is used to allow the user to provide the suggestion data for the marketing graphic materials.

5. The marketing graphic and text generation system based on workflow construction as described in claim 1 further includes a personalized record module coupled to the database module, the model building module and the graphic and text generation module, wherein the personalized record module records selection record data for the marketing graphic and text data selected by the user.

6. The marketing graphic generation system based on workflow construction as described in claim 5, wherein, The database module has a storage unit that stores the feedback data, the selection record data, and the marketing graphic data.

7. The marketing graphic generation system based on workflow construction as described in claim 6, wherein, The model building module has an optimization unit that can optimize the marketing graphic generation model based on at least one of the feedback data, the selection record data, and the marketing graphic data.

8. The marketing graphic and text generation system based on workflow construction as described in claim 1 further includes a communication module coupled to the graphic and text generation module and an interactive software of the terminal device. The user can access the startup unit and input the graphic and text request command through the interactive software and the communication module. The output unit can output the marketing graphic and text data to the interactive software through the communication module and present the marketing graphic and text data through the interactive software.