Method and system for generating images of various painting styles for specific object
The LoRA model addresses the inefficiency of existing AI models by allowing the generation of various painting styles for a specific object using a base model and additional learning, efficiently producing images that reflect the object's characteristics without needing multiple datasets, thus overcoming the resource-intensive and specialized nature of existing technologies.
Patent Information
- Application Number
- PCT/KR2024/006756
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-31
- Filing Date
- 2024-05-20
- Publication Date
- 2024-12-05
AI Technical Summary
Existing text-to-image AI models require significant resources and time for training and are specialized in specific image types, making it inefficient to generate images of various painting styles for a specific object without requiring multiple datasets.
The use of a Low Rank Adaptation (LoRA) model, which is a smaller, faster-to-train model that learns from a specific dataset based on a pre-trained stable diffusion model, allowing for the generation of images in different styles by adjusting the base model with additional learning, reducing the need for multiple datasets.
Enables the generation of images in various painting styles for a specific object using a base model and an additional learning model, efficiently producing images that reflect the characteristics of the object while minimizing resource requirements and dataset needs.
Smart Images

Figure KR2024006756_05122024_PF_FP_ABST
Abstract
Description
Method and system for generating images of various painting styles for a specific object
[0001] The following description relates to a method and system for generating images of various painting styles for a specific object.
[0002] When creating a text-to-image AI model, such as stable diffusion, the AI model is trained by inputting a large number of images and / or photographs along with descriptive text about those images and / or photographs. After training, the generated AI model file is passed to an inference tool, and prompts and parameters are specified, generating images described by the prompts. The prompts describe the desired image, which can be entered in natural language, such as "a cat jumping over a roof." Parameters can also include settings for noise methods and counts.
[0003] However, the training process of these AI models itself requires significant costs (e.g., large amounts of GPUs (Graphics Processing Units) and time). Furthermore, because the training dataset is trained to generate all types of images, including people, animals, objects, and backgrounds, it does not generate specific faces of individuals (e.g., celebrities) or specific objects (e.g., specific brands of beverages). Furthermore, because the images generated vary depending on the training data, many models specialize in generating specific images. For example, a model that trains heavily on portraits will be good at generating portraits, while a model that trains heavily on illustrations will be good at generating illustrations.
[0004] The LoRA (Low Rank Adaptation) model is a compact model trained on an additional dataset based on a specific stable diffusion model. LoRA can be an additional learning technique that trains a new subject to an existing model. These LoRA models are smaller in size, allowing them to train faster and take up less memory than stable diffusion models. However, their small size prevents them from training on a wide variety of data. A LoRA model can be created by training an image dataset (images from various angles / poses) of an object, such as a specific person, using a specific stable diffusion model as the base model. In other words, a LoRA model trained on an image dataset of an object, such as a specific person, can be generated based on a pre-trained stable diffusion model. Meanwhile, when using a LoRA model, a file for a specific stable diffusion model can be specified in an inference tool, and additionally, a specific LoRA model can be specified, such as prompts and parameters. This can generate images similar to the LoRA model's training data. In other words, images with a specified prompt feel can be obtained that are close to the dataset used to train the LoRA model.
[0005] A method and system are provided for generating images of various painting styles for a specific object using a base model and an additional learning model for a specific object.
[0006] A method for generating an image in a computer device including at least one processor, comprising: a step of designating, by the at least one processor, a first base model pre-trained to generate a picture image in a specific painting style; a step of designating, by the at least one processor, a second base model pre-trained to generate a photographic image and an additional model pre-trained to generate a photographic image of a specific object using a photograph dataset of the specific object; a step of designating, by the at least one processor, a prompt and a parameter; and a step of generating, by the at least one processor, a picture image of the specific object that includes features of the specific object and reflects the specific painting style through text-to-image conversion according to the prompt and the parameter using the first base model adjusted through the additional model.
[0007] According to one aspect, the image generation method may further include a step of designating a third base model that has been pre-trained to generate a picture image of a different style from the specific painting style; and a step of generating a picture image of the specific object that includes the characteristics of the specific object while reflecting the different painting style through text-to-image conversion according to the prompt and the parameters using the third base model adjusted through the additional model.
[0008] According to another aspect, the first base model and the second base model may include a stable diffusion model, and the additional model may be characterized in that it is generated through stable diffusion learning using the second base model and the photo dataset.
[0009] According to another aspect, the additional model may be characterized by including a LoRA (Low Rank Adaptation) model for learning the specific object in the base model.
[0010] According to another aspect, the step of generating the picture image may be characterized by including a step of adjusting the first base model so that the first base model is learned for the specific object using the additional model.
[0011] According to another aspect, the image generation method may further include a step of designating the second base model; and a step of generating a photographic image of the specific object including characteristics of the specific object through text-to-image conversion according to the prompt and the parameters using the second base model adjusted through the additional model.
[0012] A computer program stored on a computer-readable recording medium is provided for executing the above method on a computer device in combination with a computer device.
[0013] A computer-readable recording medium having recorded thereon a program for executing the above method on a computer device is provided.
[0014] A computer device is provided, comprising at least one processor implemented to execute computer-readable instructions, wherein a first base model pre-trained to generate a picture image of a specific style is specified by the at least one processor, a second base model pre-trained to generate a photographic image and an additional model pre-trained to generate a photographic image of the specific object using a photograph dataset of the specific object are specified, a prompt and a parameter are specified, and a text-to-image conversion according to the prompt and the parameter using the first base model adjusted through the additional model is performed to generate a picture image of the specific object that includes features of the specific object and reflects the specific painting style.
[0015] Using a base model and additional learning models for specific objects, we can generate images of various painting styles for a specific object.
[0016] FIG. 1 is a block diagram illustrating an example of a computer device according to one embodiment of the present invention.
[0017] FIG. 2 is a diagram illustrating an example of a process for generating (learning) a LoRA model according to an embodiment of the present invention.
[0018] FIG. 3 is a diagram illustrating an example of an image generation process using a LoRA model according to one embodiment of the present invention.
[0019] FIG. 4 is a diagram illustrating examples of actual images generated using a LoRA model trained with an actual photo dataset and various base models in one embodiment of the present invention.
[0020] FIG. 5 is a flowchart illustrating an example of an image generation method according to one embodiment of the present invention.
[0021] Hereinafter, the detailed description will be given with reference to the attached drawings.
[0022] An image generation system according to embodiments of the present invention may be implemented by at least one computer device. In this case, a computer program according to an embodiment of the present invention may be installed and run on the computer device, and the computer device may perform an image generation method according to embodiments of the present invention under the control of the run computer program. The above-described computer program may be stored on a computer-readable recording medium, which may be coupled to the computer device and cause the computer to execute the image generation method.
[0023] FIG. 1 is a block diagram illustrating an example of a computer device according to an embodiment of the present invention. As illustrated in FIG. 1, the computer device (100) may include a memory (110), a processor (120), a communication interface (130), and an input / output interface (I / O interface) 140. The memory (110) is a computer-readable recording medium and may include a random access memory (RAM), a read only memory (ROM), and a permanent mass storage device such as a disk drive. Here, the ROM and the permanent mass storage device such as the disk drive may be included in the computer device (100) as a separate permanent storage device distinct from the memory (110). In addition, an operating system and at least one program code may be stored in the memory (110). These software components may be loaded into the memory (110) from a computer-readable recording medium separate from the memory (110). This separate computer-readable recording medium may include a computer-readable recording medium such as a floppy drive, a disk, a tape, a DVD / CD-ROM drive, a memory card, etc. In another embodiment, the software components may be loaded into the memory (110) via a communication interface (130) other than a computer-readable recording medium. For example, the software components may be loaded into the memory (110) of the computer device (100) based on a computer program installed by files received via a network (Network, 160).
[0024] The processor (120) may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. Instructions may be provided to the processor (120) via the memory (110) or the communication interface (130). For example, the processor (120) may be configured to execute instructions received according to program code stored in a storage device such as the memory (110).
[0025] The communication interface (130) may provide a function for the computer device (100) to communicate with other devices via a network (160). For example, requests, commands, data, files, etc. generated by the processor (120) of the computer device (100) according to program codes stored in a recording device such as a memory (110) may be transmitted to other devices via the network (160) under the control of the communication interface (130). Conversely, signals, commands, data, files, etc. from other devices may be received by the computer device (100) via the communication interface (130) of the computer device (100) via the network (160). Signals, commands, data, etc. received via the communication interface (130) may be transmitted to the processor (120) or the memory (110), and files, etc. may be stored in a storage medium (the aforementioned permanent storage device) that the computer device (100) may further include.
[0026] The input / output interface (140) may be a means for interfacing with an input / output device (I / O device, 150). For example, the input device may include a device such as a microphone, a keyboard, or a mouse, and the output device may include a device such as a display or a speaker. As another example, the input / output interface (140) may be a means for interfacing with a device that integrates input and output functions, such as a touchscreen. The input / output device (150) may also be configured as a single device with the computer device (100).
[0027] Additionally, in other embodiments, the computer device (100) may include fewer or more components than those illustrated in FIG. 1. However, it is not necessary to explicitly illustrate most conventional components. For example, the computer device (100) may be implemented to include at least some of the input / output devices (150) described above, or may further include other components such as a transceiver, a database, etc.
[0028] FIG. 2 is a diagram illustrating an example of a process for generating (learning) a LoRA model according to an embodiment of the present invention. FIG. 2 shows an example in which a pre-trained stable diffusion model A (210) is designated as a base model. At this time, it is assumed that the stable diffusion model A (210) has been pre-trained to generate a photographic image. As training data, a photographic dataset (220) of a specific object (e.g., a specific person) can be used. It is assumed that the photographic dataset (220) includes images of various angles and / or poses for the specific object. The embodiment of FIG. 2 shows that a LoRA model X (240) can be generated through a process of stable diffusion learning (230) using the stable diffusion model A (210) designated as a base model and the photographic dataset (220). Since the technology itself for stable diffusion learning for generating a LoRA model is already well known, a detailed description thereof will be omitted. In this embodiment, we focus on creating a LoRA model X (240) using a pre-trained stable diffusion model A (210) to generate a photo image and a photo dataset (220) for a specific object.
[0029] FIG. 3 is a diagram illustrating an example of an image generation process using a LoRA model according to an embodiment of the present invention. FIG. 3 shows an embodiment in which a pre-trained stable diffusion model B (310) is used as a base model for stable diffusion text-to-image conversion (320) to generate an illustration image of a first style. At this time, the stable diffusion text-to-image conversion (320) may generate an image (350) by designating a file of the stable diffusion model B (310) to an inference tool, and designating a prompt (330), a parameter (340), and a LoRA model X (240) generated through the process of FIG. 2. At this time, the generated image (350) may include a picture image that appears to have been drawn in the first style of the stable diffusion model B (310) while maintaining the characteristics of the photo dataset (220) of a specific object used for training the LoRA model X (240). In other words, the characteristics of the object can be generated as an image (350) that follows the characteristics of the photograph dataset (220) of the LoRA model X (240), while the picture image following the first painting style of the stable diffusion model B (310) can be generated. In this case, it becomes possible to change the painting style by replacing the base model. For example, when using a pre-trained base model 1 to generate an illustration image of a second painting style (e.g., watercolor style), the characteristics of the object can be generated as an illustration image drawn in the second painting style while following the characteristics of the photograph dataset (220) of the LoRA model X (240). As another example, when using a pre-trained base model 2 to generate an illustration image of a third painting style (e.g., pencil drawing style), the characteristics of the object can be generated as an illustration image drawn in the third painting style while following the characteristics of the photograph dataset (220) of the LoRA model X (240).As another example, if the pre-trained base model 3 is used to generate a photo image as a base model, a photo image can be generated whose characteristics follow the characteristics of the photo dataset (220) of the LoRA model X (240).
[0030] In this way, by changing the base model, it becomes possible to obtain various painting images of various painting styles related to a specific object (e.g., a specific person) trained with photographic images. That is, in the past, in order to generate a painting image of a specific object, multiple painting images of the object were required as training images. On the other hand, in the embodiments of the present invention, painting images of a specific painting style for a specific object can be obtained by using a LoRA model trained with photographic images of a specific object and a base model trained with painting images of a specific painting style rather than a specific object, thereby eliminating the need to obtain multiple painting datasets for a specific object. In addition, painting images of various painting styles for a specific object can be obtained without requiring multiple painting datasets for a specific object.
[0031] FIG. 4 is a diagram illustrating examples of actual images generated using a LoRA model trained with an actual photo dataset and various base models according to an embodiment of the present invention. The embodiment of FIG. 4 shows an example of a LoRA model (420) generated using an actual photo dataset (410) of a specific person A according to the learning method described in the embodiment of FIG. 2 above. Here, the LoRA model (420) of FIG. 4 may be a model trained to generate a photo image of a specific person A.
[0032] At this time, even if the LoRA model (420) trained with the actual photo dataset (410) for a specific person A is used, different results may be derived depending on which model is designated as the base model. In the embodiment of Fig. 4, an example is shown in which a photo image (460) for a specific person A is generated by designating the pre-trained stable diffusion model 1 (430) to generate a photo image as the base model. In addition, in the embodiment of Fig. 4, an example is shown in which a picture image (470) as a watercolor-style illustration image for a specific person A is generated by designating the pre-trained stable diffusion model 2 (440) to generate a watercolor-style illustration image as the base model. In addition, in the embodiment of Fig. 4, an example is shown in which a picture image (480) as a pencil-drawing style illustration image for a specific person A is generated by designating the pre-trained stable diffusion model 3 (450) to generate a pencil-drawing style illustration image as the base model.
[0033] In this way, it becomes possible to generate various painting images of a specific person A in various painting styles without having to obtain various painting images of a specific person A as learning data.
[0034] FIG. 5 is a flowchart illustrating an example of an image generation method according to an embodiment of the present invention. The image generation method according to the present embodiment can be performed by a computer device (100). At this time, the processor (120) of the computer device (100) can be implemented to execute control instructions according to the code of the operating system included in the memory (110) or the code of at least one computer program. Here, the processor (120) can control the computer device (100) so that the computer device (100) performs steps (510 to 580) included in the method of FIG. 5 according to the control instructions provided by the code stored in the computer device (100).
[0035] In step (510), the computer device (100) may be assigned a first base model that has been pre-trained to generate a picture image of a specific painting style. For example, the first base model may include, but is not limited to, a stable diffusion model. For example, the first base model may be an artificial intelligence model capable of generating a picture image of a specific painting style through text-to-image conversion, and capable of learning about a specific object through an additional model described later, and may be used without limitation. As examples of painting styles, watercolor and pencil drawing styles have been described above, but painting styles may include various types that can be classified, including meanings such as painting style or drawing style. For example, various types of painting styles, painting styles, and drawing styles, such as a feminine deformed painting style, a community style, a Twitter style, a CalArts style, and a dramatic style, may be included in the painting styles according to embodiments of the present invention.
[0036] In step (520), the computer device (100) may be assigned a second base model pre-trained to generate a photographic image and an additional model pre-trained to generate a photographic image of a specific object using a photographic dataset of the specific object. The second base model may also include a stable diffusion model. This second base model may be pre-trained to generate a photographic image, and the additional model may be trained using a photographic dataset reflecting various angles and / or poses of the specific object to generate a photographic image of the specific object based on this second base model. This additional model may include, but is not limited to, a LoRA model for training the base model for a specific object. For example, a model generated by training the second base model pre-trained to generate a photographic image for a specific object to generate a photographic image of the specific object may be utilized as the additional model.
[0037] At step (530), the computer device (100) may be assigned prompts and parameters. The prompt may be a description of the image to be generated, which may be entered in natural language, such as "a cat jumping over a roof." Furthermore, the parameters may include settings for noise methods, counts, etc.
[0038] Steps (510), (520), and (530) may be processes for assigning a first base model, an additional model, a prompt, and parameters to the inference tool, respectively, and the order thereof may be changed. For example, the prompts and parameters may be assigned before the additional model, or the first base model may be assigned first. Here, the processes for assigning the first base model, an additional model, a prompt, and parameters to the inference tool, respectively, may include processes for inputting files, data, or setting values for assignment to the inference tool.
[0039] In step (540), the computer device (100) can generate a pictorial image of a specific object that reflects a specific drawing style while including the characteristics of the specific object through text-to-image conversion according to prompts and parameters using the first base model adjusted through the additional model. In this case, the computer device (100) can adjust the first base model using the additional model so that the first base model is trained on the specific object. Through this, the first base model can generate a pictorial image that includes the characteristics of the specific object while having a specific drawing style.
[0040] Additionally, the computer device (100) may generate a drawing image of a specific object in a different style. To this end, the computer device (100) may perform steps (550) and (560).
[0041] In step (550), the computer device (100) may be assigned a third base model pre-trained to generate a picture image of a specific style and a different style. Here, the third base model may also include, but is not limited to, a stable diffusion model. Furthermore, assigning the third base model may mean assigning the third base model to replace the first base model assigned in step (510).
[0042] In step (560), the computer device (100) can generate a pictorial image of a specific object that reflects different drawing styles while including the characteristics of the specific object through text-to-image conversion according to prompts and parameters using a third base model adjusted through an additional model. In other words, if there are various base models that have been pre-trained to generate pictorial images of various drawing styles, the computer device (100) can generate pictorial images of various drawing styles that reflect the drawing styles of each base model for the specific object. Considering that it is more difficult to obtain a drawn image of a specific object than a photographic image of the specific object, the computer device (100) has the advantage of being able to generate pictorial images of a specific object in various drawing styles without acquiring a learning pictorial image of the specific object.
[0043] Additionally, the computer device (100) may generate a photographic image of a specific object if it utilizes a second base model for generating a photographic image. To this end, the computer device (100) may perform steps (570) and (580).
[0044] In step (570), the computer device (100) may be assigned a second base model. Here, the second base model may refer to a base model for learning the additional model previously assigned in step (520). In addition, assigning the second base model may refer to assigning the second base model as a replacement for the first base model or the third base model previously assigned in step (510).
[0045] At step (580), the computer device (100) can generate a photographic image of a specific object containing the characteristics of the specific object through text-to-image conversion according to prompts and parameters using a second base model adjusted through an additional model. In other words, depending on the selection of the base model, not only various painting styles of the specific object but also photographic images of the specific object can be obtained.
[0046] In this way, according to embodiments of the present invention, images of various painting styles for a specific object can be generated using a base model and an additional learning model for the specific object.
[0047] The systems or devices described above may be implemented as hardware components, or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding to them. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing device is sometimes described as being used alone; however, one of ordinary skill in the art will recognize that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processing unit may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.
[0048] Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing device to perform a desired operation or may independently or collectively command the processing device. The software and / or data may be embodied in any type of machine, component, physical device, virtual equipment, computer storage medium, or device for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.
[0049] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., singly or in combination. The medium may continuously store a computer-executable program, or may temporarily store it for execution or download. In addition, the medium may be various recording means or storage means in the form of a single or multiple hardware combinations, and is not limited to a medium directly connected to a computer system, but may also be distributed over a network. Examples of the medium may include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and those configured to store program commands, including ROM, RAM, and flash memory. In addition, examples of other media may include recording media or storage media managed by app stores that distribute applications, sites that supply or distribute various software, servers, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.
[0050] Although the embodiments described above have been described by way of limited examples and drawings, those skilled in the art will appreciate that various modifications and variations can be made based on the above teachings. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.
[0051] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.
Claims
1. A computer program stored in a computer-readable recording medium for executing an image generation method on a computer device by being combined with a computer device, The above image generation method is, A step of specifying a first base model that is pre-trained to generate a painting image of a specific painting style; A step of designating a second base model pre-trained to generate a photographic image and an additional model pre-trained to generate a photographic image of a specific object using a photographic dataset of the specific object; Steps for specifying prompts and parameters; and A step of generating a picture image of the specific object that reflects the specific drawing style while including the characteristics of the specific object through text-to-image conversion according to the prompt and the parameters using the first base model adjusted through the additional model. Including A computer program characterized by .
2. In paragraph 1, The above image generation method is, A step of designating a third base model that has been pre-trained to generate a painting image of a different painting style from the above-mentioned specific painting style; and A step of generating a picture image of the specific object that includes the characteristics of the specific object and reflects the different drawing style through text-to-image conversion according to the prompt and the parameters using the third base model adjusted through the additional model. Including more A computer program characterized by .
3. In paragraph 1, The first base model and the second base model include a stable diffusion model, The above additional model is created through stable diffusion learning using the second base model and the photo dataset. A computer program characterized by .
4. In paragraph 1, The above additional model includes a LoRA (Low Rank Adaptation) model for learning the specific object in the base model. A computer program characterized by .
5. In paragraph 1, The steps for generating the above picture image are: A step of adjusting the first base model so that the first base model is learned for the specific object using the additional model. Including A computer program characterized by .
6. In paragraph 1, The above image generation method is, A step of designating the second base model; and A step of generating a photographic image of the specific object including the characteristics of the specific object through text-to-image conversion according to the prompt and the parameters using the second base model adjusted through the additional model. Including more A computer program characterized by .
7. A method for generating an image of a computer device including at least one processor, A step of designating a first base model pre-trained to generate a picture image of a specific style by at least one processor; A step of designating, by at least one processor, a second base model pre-trained to generate a photographic image and an additional model pre-trained to generate a photographic image of a specific object using a photographic dataset of the specific object; A step of specifying a prompt and parameters by at least one processor; and A step of generating a picture image of the specific object that includes the characteristics of the specific object and reflects the specific drawing style through text-to-image conversion according to the prompt and the parameters using the first base model adjusted through the additional model by at least one processor. A method for generating an image including:
8. In paragraph 7, A step of designating a third base model pre-learned to generate a picture image of a different style from the specific style by at least one processor; and A step of generating a picture image of the specific object that includes the characteristics of the specific object and reflects the different drawing style through text-to-image conversion according to the prompt and the parameters using the third base model adjusted through the additional model by at least one processor. A method for generating an image that further includes .
9. In paragraph 7, The first base model and the second base model include a stable diffusion model, The above additional model is created through stable diffusion learning using the second base model and the photo dataset. A method for generating an image characterized by .
10. In paragraph 7, An image generation method characterized in that the above additional model includes a LoRA (Low Rank Adaptation) model for learning the specific object in the base model.
11. In paragraph 7, The steps for generating the above picture image are: A step of adjusting the first base model so that the first base model is learned for the specific object using the additional model. A method for generating an image, characterized in that it includes:
12. In paragraph 7, A step of designating the second base model by at least one processor; and A step of generating a photographic image of the specific object including features of the specific object through text-to-image conversion according to the prompt and the parameters using the second base model adjusted through the additional model by the at least one processor. A method for generating an image that further includes .
13. A computer-readable recording medium having recorded thereon a computer program for executing the method of any one of claims 7 to 12 on a computer device.
14. At least one processor implemented to execute computer-readable instructions. Including, By at least one processor, A first base model is given that is pre-trained to generate a painting image of a specific style, A second base model pre-trained to generate a photographic image and an additional model pre-trained to generate a photographic image of a specific object using a photographic dataset of the specific object are specified, Given prompts and parameters, Generating a picture image of the specific object that reflects the specific drawing style while including the characteristics of the specific object through text-to-image conversion according to the prompt and the parameters using the first base model adjusted through the additional model. A computer device characterized by:
15. In paragraph 14, By at least one processor, A third base model that is pre-trained to generate a painting image of a different style from the above-mentioned specific painting style is specified, Generating a picture image of the specific object that includes the characteristics of the specific object and reflects the different drawing style through text-to-image conversion according to the prompt and the parameters using the third base model adjusted through the additional model by at least one processor A computer device characterized by:
Citation Information
Patent Citations
Identifying digital data with a new style
JP7270894B1
Method for preparing composite solid electrolyte
KR1020230166967A
An Apparatus for Protecting a Drill
KR1020240174384A
Artificial intelligence content creation system and method
KR102313203B1
Low-Rank Adaptation of Neural Network Models
US20220383126A1
Cited By
Building video generation method based on natural language and computer program product
CN119919518A
