Converting images into vector graphics

Abstract

The present disclosure describes techniques for converting images created using generative artificial intelligence, to one or more vector graphics. Generative artificial intelligence may create one or more images in response to receiving a prompt. The one or more images may be converted into one or more vector graphics, which may be reproduced on one or more products using laser etching or printing techniques.

Description

PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136CONVERTING IMAGES INTO VECTOR GRAPHICSCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to and benefit of U.S. Patent Application No. 63 / 707,424 filed on October 15, 2024. The contents of the above listed application are incorporated herein by reference in their entirety for any and all non-limiting purposes.FIELD OF THE INVENTION

[0002] Aspects of the disclosure generally relate to prompt engineering and, more specifically, generating one or more images in response to receiving a prompt and converting the one or more images to vector graphics.BACKGROUND OF THE INVENTION

[0003] Generative artificial intelligence may be used to produce images. However, generative artificial intelligence does not produce useful vector graphics. In this regard, the images produced by generative artificial intelligence oftentimes include too much detail that cannot be reproduced.SUMMARY OF THE INVENTION

[0004] The following presents a simplified summary of various features described herein. This summary is not an extensive overview, and is not intended to identify key or critical elements or to delineate the scope of the claims. The following summary merely presents some concepts in a simplified form as an introductory prelude to the more detailed descriptions provided below. Corresponding apparatus, systems, and computer-readable media are also within the scope of the disclosure.

[0005] Aspects of the disclosure generally relate to prompt engineering. Additional aspects of the disclosure generally relate to converting one or more images, created using generative artificial intelligence, to vector graphics.

[0006] The prompt engineering described herein may improve the ability of generative artificial intelligence to create (e.g., produce, generate, render, etc.) vector graphics (e.g., line drawings). In this regard, a computing device may receive an input via a prompt. The input may include a request to generate one or more images. The request may comprise a description, or details, to include in the one or more images. ToPCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 improve the ability of generative artificial intelligence to create (e.g., produce, generate, render, etc.) vector graphics, the prompt may cause one or more suggestions to be displayed to improve the results. Additionally, the prompt, or a back-end function of the prompt, may modify the input, for example, by adding one or more additional conditions, terms, and / or phrases to produce better results that are either vector graphics or could easily be converted into vector graphics.

[0007] Regardless of whether the input has been modified, the input may be provided to a generative artificial intelligence model. In this regard, the prompt may be part of an interface provided by a computing device. The interface may transmit (e.g., send) the input to the generative artificial intelligence model, for example, via one or more application programming interfaces (APIs). Upon receiving the input, the generative artificial intelligence model may create (e.g., produce, generate, render, etc.) one or more images based on the input and based on any additional terms added to the input. The generative artificial intelligence model may transmit (e.g., send) the one or more images to the computing device. The computing device may review the one or more images created by the generative artificial intelligence model to determine whether the one or more images comply with one or more policies. The one or more policies may restrict images that are lewd, offensive, use foul language, include explicit material and / or language, infringe a third party’s intellectual property rights (e.g., copyrights, trademarks, etc.), and the like. The computing device may cause a subset of images that comply with the one or more policies to be converted into vector graphics. Each of the vector graphics associated with each of the images, of the subset of images, may be displayed via the interface. By converting the images to vector graphics, the present disclosure addresses a shortcoming of existing generative artificial intelligence models.

[0008] These features, along with many others, are discussed in greater detail below.BRIEF DESCRIPTION OF DRAWINGS

[0009] The present disclosure is described by way of example and not limited in the accompanying figures in which like reference numerals indicate similar elements and in which:

[0010] FIG. 1 shows an example of a system in which one or more features described herein may be implemented;PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136

[0011] FIG. 2 shows an example of a computing device in accordance with one or more aspects of the disclosure;

[0012] FIG. 3 shows an example of a process for converting images, created using generative artificial intelligence, to vector graphics in accordance with one or more aspects of the disclosure; and

[0013] FIGS. 4A-4D show an example of an interface for creating images using one or more generative artificial intelligence models in accordance with one or more aspects of the disclosure.DETAILED DESCRIPTION OF THE INVENTION

[0014] In the following description, reference is made to the accompanying drawings, which form a part hereof, and in which are shown various examples of features of the disclosure and / or of how the disclosure may be practiced. It is to be understood that other features may be utilized and structural and functional modifications may be made without departing from the scope of the present disclosure. The disclosure may be practiced or carried out in various ways. In addition, it is to be understood that the phraseology and terminology used herein are for the purpose of description and should not be regarded as limiting. Rather, the phrases and terms used herein are to be given their broadest interpretation and meaning consistent with the disclosures and / or descriptions below.

[0015] By way of introduction, features discussed herein may relate to prompt engineering to generate one or more images using generative artificial intelligence models. Further aspects of the disclosure relate to converting one or more images, created using one or more generative artificial intelligence models, to vector graphics. The vector graphics may be etched (e.g., laser etched) or printed on products, such as drmkware, to customize the products.

[0016] As discussed above, generative artificial intelligence models struggle to produce vector graphics. In this regard, the present disclosure employs prompt engineering to create and / or refine inputs to generative artificial intelligence models such that the generative artificial intelligence models produce images that are better-suited to be converted to vector graphics (e.g., line drawings). A computing device may cause an interface to be displayed on a user device (e.g., mobile device, a web browser on a laptop, etc.). The interface may comprise a prompt configured to receive an input. The input may include a request to generate one or more images. The input mayPCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 comprise a description of the subject matter of the one or more images. Additionally or alternatively, the input may comprise details to include in the one or more images. In some instances, the prompt may cause one or more suggestions to be displayed to improve the results generated by the generative artificial intelligence model. Additionally or alternatively, the computing device may modify the input prior to providing the input to the generative artificial intelligence model. The computing device may modify the input, for example, by adding additional terms and / or conditional statements to the input. In some instances, the additional terms and / or conditional statements may comprise negative conditions.

[0017] After receiving the input, the computing device may transmit (e.g., send) the input to the generative artificial intelligence model. The generative artificial intelligence model may create (e.g., produce, generate, render, etc.) one or more images based on the input and / or based on any additional terms and / or conditional statements added to the input. The generative artificial intelligence model may transmit (e.g., send) the one or more images, generated by the generative artificial intelligence model, to the computing device. The computing device may review the one or more images created by the generative artificial intelligence model to determine whether the one or more images comply with one or more policies. The one or more policies may prevent images that are lewd, offensive, use foul language, include explicit material and / or language, infringe a third party's intellectual property rights (e.g., copyrights, trademarks, etc.), and the like, from being displayed. After filtering the images, the computing device may cause a subset of images to be converted into vector graphics. The computing device may cause each of the vector graphics associated with each of the images, of the subset of images, to be displayed via the interface. In response to causing the vector graphics to be displayed, the computing device may receive a selection of a first vector graphic. The computing device may cause the selected first vector graphic to displayed, via the interface, as an overlay on a product, such as drinkware. Additionally or alternatively, the computing device may cause the selected first vector graphic to be reproduced on the product.

[0018] FIG. 1 shows an example of a system 100 that includes a first user device 110, a second user device 120, and a server 130, connected to a first database 140, interconnected via network 150.

[0019] First user device 110 may be a mobile device, such as a cellular phone, a mobile phone, a smart phone, a tablet, a laptop, or an equivalent thereof. First user devicePCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136110 may be associated with a first user. First user device 110 may provide a first user with access to various applications and / or services, including one or more applications and / or services that allow the first user to access the Internet. Additionally, first user device 110 may provide the first user with one or more applications C apps") located thereon. The one or more applications may provide the first user with a plurality of tools and access to a variety of services.

[0020] Second user device 120 may be a computing device configured to allow a user to execute software for a variety of purposes. Second user device 120 may belong to the first user that accesses first user device 110, or, alternatively, second user device 120 may belong to a second user, different from the first user. Second user device 120 may be a desktop computer, laptop computer, or, alternatively, a virtual computer. The software of second user device 120 may include one or more web browsers that provide access to websites on the Internet. The one or more web browsers may allow the second user to access a webpage that allows users to customize products, such as drinkware. The second user may log in to the webpage and customize one or more products for purchase.

[0021] Server 130 may be any server capable of executing application 132. In this regard, server 130 may be a stand-alone server, a corporate server, or a server located in a server farm or cloud-computer environment. According to some examples, server 130 may be a virtual server hosted on hardware capable of supporting a plurality of virtual servers. Additionally, server 130 may be communicatively coupled to first database 140.

[0022] Application 132 may be server-based software configured to allow users to customize products. Application 132 may be server-based software associated with client-based software executing on the user devices described above (e.g., first user device 110, second user device 120, etc.). Application 132 may provide one or more webpage configured to display products. Additionally or alternatively, the one or more webpage may allow for customization of the products.

[0023] First database 140 may be configured to store information on behalf of application 132. The information may include, but is not limited to, user information, product information, etc. First database 140 may include, but is not limited to relational databases, hierarchical databases, distributed databases, in-memory databases, flat file databases, XML databases, NoSQL databases, graph databases, and / or a combination thereof.PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136

[0024] Network 150 may include any type of network, including the Internet, a local area network (LAN), a wide area network (WAN), a wireless telecommunications network, and / or any other communication network or combination thereof. It will be appreciated that the network connections shown are illustrative and any means of establishing a communications link between the computers may be used. The existence of any of various network protocols such as TCP / IP, Ethernet, FTP, HTTP and the like, and of various wireless communication technologies such as GSM, CDMA, WiFi, and LTE, is presumed, and the various computing devices described herein may be configured to communicate using any of these network protocols or technologies. The data transferred to and from various computing devices in system 100 may include secure and sensitive data, such as confidential documents, customer personally identifiable information, and account data. Therefore, it may be desirable to protect transmissions of such data using secure network protocols and encryption, and / or to protect the integrity of the data when stored on the various computing devices. For example, a file-based integration scheme or a service-based integration scheme may be utilized for transmitting data between the various computing devices. Data may be transmitted using various network communication protocols. Secure data transmission protocols and / or encryption may be used in file transfers to protect the integrity of the data, for example, File Transfer Protocol (FTP), Secure File Transfer Protocol (SFTP), and / or Pretty Good Privacy (PGP) encryption. In many embodiments, one or more web services may be implemented within the various computing devices. Web services may be accessed by authorized external devices and users to support input, extraction, and manipulation of data between the various computing devices in the system 100. Web services built to support a personalized display system may be cross-domain and / or cross-platform, and may be built for enterprise use. Data may be transmitted using the Secure Sockets Layer (SSL) or Transport Layer Security (TLS) protocol to provide secure connections between the computing devices. Web services may be implemented using the WS-Security standard, providing for secure SOAP messages using encryption. Specialized hardware may be used to provide secure web services. For example, secure network appliances may include built-in features such as hardware-accelerated SSL and HTTPS, WS-Security, and / or firewalls. Such specialized hardware may be installed and configured in system 100 in front of one or more computing devices such that any external devices may communicate directly with the specialized hardware.PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136

[0025] Any of the devices and systems described herein may be implemented, in whole or in part, using one or more computing devices described with respect to FIG. 2. Turning now to FIG. 2, a computing device 200 that may be used with one or more of the computational systems is described. The computing device 200 may comprise a processor 203 for controlling overall operation of the computing device 200 and its associated components, including RAM 205, ROM 207, input / output device 209, accelerometer 211, global-position system antenna 213, memory 215, and / or communication interface 223. A bus 202 may interconnect processor(s) 203, RAM 205, ROM 207, memory 215, I / O device 209, accelerometer 211, global-position system receiver / antenna 213, memory 215, and / or communication interface 223. Computing device 200 may represent, be incorporated in, and / or comprise various devices such as a desktop computer, a computer server, a gateway, a mobile device, such as a laptop computer, a tablet computer, a smart phone, any other types of mobile computing devices, and the like, and / or any other type of data processing device.

[0026] Input / output (I / O) device 209 may comprise a microphone, keypad, touch screen, and / or stylus through which a user of the computing device 200 may provide input, and may also comprise one or more of a speaker for providing audio output and a video display device for providing textual, audiovisual, and / or graphical output. Software may be stored within memory 215 to provide instructions to processor 203 allowing computing device 200 to perform various actions. For example, memory 215 may store software used by the computing device 200, such as an operating system 217, application programs 219, and / or an associated internal database 221. The various hardware memory units in memory 215 may comprise volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. Memory 215 may comprise one or more physical persistent memory devices and / or one or more non-persistent memory devices. Memory 215 may comprise random access memory (RAM) 205, read only memory (ROM) 207, electronically erasable programmable read only memory (EEPROM), flash memory or other memory technology, optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that may be used to store the desired information and that may be accessed by processor 203.PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136

[0027] Accelerometer 211 may be a sensor configured to measure accelerating forces of computing device 200. Accelerometer 211 may be an electromechanical device. Accelerometer may be used to measure the tilting motion and / or orientation computing device 200, movement of computing device 200, and / or vibrations of computing device 200. The acceleration forces may be transmitted to the processor to process the acceleration forces and determine the state of computing device 200.

[0028] GPS receiver / antenna 213 may be configured to receive one or more signals from one or more global positioning satellites to determine a geographic location of computing device 200. The geographic location provided by GPS receiver / antenna 213 may be used for navigation, tracking, and positioning applications. In this regard, the geographic may also include places and routes frequented by the first user.

[0029] Communication interface 223 may comprise one or more transceivers, digital signal processors, and / or additional circuitry and software, protocol stack, and / or network stack for communicating via any network, wired or wireless, using any protocol as described herein.

[0030] Processor 203 may comprise a single central processing unit (CPU), which may be a single-core or multi-core processor, or may comprise multiple CPUs. Processors) 203 and associated components may allow the computing device 200 to execute a series of computer-readable instructions (e.g., instructions stored in RAM 205, ROM 207, memory 215, and / or other memory of computing device 215, and / or in other memory) to perform some or all of the processes described herein. Although not shown in FIG. 2, various elements within memory 215 or other components in computing device 200, may comprise one or more caches, for example, CPU caches used by the processor 203, page caches used by the operating system 217, disk caches of a hard drive, and / or database caches used to cache content from database 221. A CPU cache may be used by one or more processors 203 to reduce memory latency and access time. A processor 203 may retrieve data from or write data to the CPU cache rather than reading / writing to memory 215, which may improve the speed of these operations. In some examples, a database cache may be created in which certain data from a database 221 is cached in a separate smaller database in a memory separate from the database, such as in RAM 205 or on a separate computing device. For example, in a multi-tiered application, a database cache on an application server may reduce data retrieval and data manipulation time by not needing to communicate over a network with a back-end database server. These types of caches and othersPCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 may provide potential advantages in certain implementations of devices, systems, and methods described herein, such as faster response times and less dependence on network conditions when transmitting and receiving data.

[0031] Although various components of computing device 200 are described separately, functionality of the various components may be combined and / or performed by a single component and / or multiple computing devices in communication without departing from the disclosure.

[0032] As noted above, the use of generative artificial intelligence models is becoming more and more common. However, images created using generative artificial intelligence models are, oftentimes, ill-suited for reproduction, especially when being reproduced on products, such as drinkware. FIG. 3 shows an example of a process for converting images, created using one or more generative artificial intelligence models, to vector graphics in accordance with one or more aspects of the disclosure. Some or all of the steps of the process shown in FIG. 3 may be performed using one or more computing devices as descnbed herein, including, for example first user device 110, second user device 120, and / or server 130.

[0033] In step 305, a computing device may receive an input via a prompt of an interface. The computing device may be a server, similar to server 130 discussed above. The server may cause the interface to be displayed on a user device, such as first user device 110 and / or second user device 120. The interface may be a mobile application executing on the user device. Additionally or alternatively, the interface may be a webpage, or part of a webpage, that is displayed via a web browser or a mobile application executing on the user device. The prompt may be text-based. That is, the prompt may request a written (e.g., typed) description of the user’s request. Additionally or alternatively, the prompt may be speech-based. In this regard, the prompt may activate an input device, such as a microphone and / or a camera, of the user device to allow the user to describe the request via speech. The user device may convert the spoken input to text via a speech-to-text algorithm. In further examples, the prompt may be an image. In this regard, the user may be able to upload an image and / or a photo via the prompt. Additionally or alternatively, the user may be able to activate a camera, or other input device, on their user device to capture (e.g., obtain) an image for the input.

[0034] In step 310, the computing device may analyze the input to determine whether the input complies with the one or more policies. The input may be analyzed prior to thePCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 input being provided to a generative artificial intelligence model. In some instances, the analysis may be performed by one or more machine learning models, different from the generative artificial intelligence model. The one or more machine learning models may be trained to identify objectionable inputs. The one or more machine learning models may comprise neural networks, such as large language models (LLM(s)), convolutional neural networks (CNN), recurrent neural networks, recursive neural networks, long short-term memory (LSTM), gated recurrent units (GRU), unsupervised pre-trained networks, space invariant artificial neural networks, generative adversarial networks (GAN), consistent adversarial networks (CAN), such as cyclic generative adversarial networks (C-GAN), deep convolutional GANs (DC- GAN), GAN interpolation (GAN-INT), GAN-CLS, a cyclic-CAN (e.g., C-CAN), or any equivalent and / or combination thereof. Additionally or alternatively, the one or more machine learning models may comprise one or more decision trees. The one or more machine learning models may be trained using supervised learning, unsupervised learning, back propagation, transfer learning, Adam stochastic optimization, stochastic gradient descent, learning rate decay, dropout, max pooling, batch normalization, long short-term memory, skip-gram, or any equivalent deep learning technique. The one or more machine learning models may be trained, for example, using the one or more policies. Additionally or alternatively, the one or more machine learning models may be trained using objectionable and / or protected data. A corpus of the training data set may be divided into training data and testing data. Preferably, 65% to 85% of the corpus would form the training data, while the remaining 15% to 35% of the corpus would be test data. The one or more machine learning models may be trained using the training data, while the test data would be used to help the one or more machine learning models achieve convergence (i.e., an error range with an acceptable tolerance).

[0035] The objectionable inputs may be defined by the one or more policies, which may indicate acceptable and unacceptable terms and / or phrases. For example, the one or more policies may prohibit terms and / or phrases that may be lewd, offensive, sexist, racist, misogynistic, etc. Additionally or alternatively, the one or more policies may prohibit inputs directed to trademarks or copyrightable works. For instance, the one or more policies may prohibit an input that includes a cartoon character. In another example, the one or more policies may prohibit an input that includes a team name or logo. If the computing device determines that the input does not comply with the one ioPCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 or more policies, the computing device may cause an error message to be displayed on the user device, in step 313. The error message may comprise one or more instructions to revise the second input. In this regard, the error message may provide feedback to a user with respect to structuring an instruction that can be interpreted and understood by the generative artificial intelligence model and that complies with the one or more policies.

[0036] After receiving an input that complies with the one or more policies, the computing device may modify the input, in step 315. Modifying the input may comprise adding one or more conditions to the input. The one or more conditions may provide additional details with respect to the requested output. For example, the one or more conditions may request that the output be black-and-white and / or a digital vector graphic. Additionally, the one or more conditions may request that the output be centered, specify a background color for the image (e.g., white), and / or that the output be a stylized logo without words. In some instances, the one or more conditions may comprise negative conditions. That is, the modifications may specify that the output not have thin lines, not be gray, not have shading, not have a dark background (e.g., black), not have text, not have words, not have details, not have fur, not have hair, not be photorealistic, not include a picture frame, not be a photo, not be a picture, etc. It will be appreciated that these modifications are merely illustrative and additional conditions may be applied. It will be further appreciated that any combination of the conditions specified above may be used in combination to obtain a desired outcome.

[0037] Based on the input and any modifications applied thereto, the computing device may generate a plurality of images, for example, using one or more generative artificial intelligence models. In step 320, the computing device may transmit (e.g., send) the input, including any modifications made thereto, to the one or more generative artificial intelligence models. In some instances, the input may be transmitted to the one or more generative artificial intelligence models via one or more application programming interfaces (APIs). In other words, the interface may receive the input, entered via a user device, and send the input to a back-end generative artificial intelligence model, executing on the computing device, via the one or more APIs. The generative artificial intelligence model may be configured to create (e.g., generate, produce, etc.) one or more images based on the received input. In this regard, the generative artificial intelligence model may be any suitable generative artificial intelligence model, including, for example, Imagen Al, Google® Vertex Al,PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136Midjoumey, OpenAI's DALL-E, Stability Al's Stable Diffusion, etc. Additionally or alternatively, the generative artificial intelligence model may be a proprietary generative artificial intelligence model trained to create (e.g., generate, produce, etc.) one or more images based on a received input.

[0038] In step 325, the computing device may receive a plurality of images from the one or more generative artificial intelligence models. The plurality of images may be received via the one or more APIs discussed above. The plurality of images may be based on the input received via the prompt. In step 330, the computing device may analyze (e g. moderate) the plurality of images to determine whether the plurality of images complies with one or more policies. The plurality of images may be analyzed using one or more machine learning models. The one or more policies may be the same, or similar, to the one or more policies discussed above. Similarly, the one or more machine learning models may be the same, or similar, to the one or more machine learning models discussed above. Alternatively, the one or more policies and / or the one or more machine learning models used to analyze (e.g., moderate) the plurality of images may be different from the one or more policies and / or the one or more machine learning models used to analyze the input. In this regard, the one or more machine learning models used to analyze (e.g., moderate) the plurality of images may be one or more machine learning models trained to analyze images, whereas the one or more machine learning models used to analyze (e.g., moderate) the input may be one or more machine learning models trained to analyze and / or review text input.

[0039] In step 335, the computing device may select a subset of images, from the plurality of images, to be displayed via the user device. The selection of the subset of images may be based on a determination that the subset of images complies with the one or more policies. In other words, the computing device may dispose of images that do not comply with the one or more policies. For example, the computing device may dispose of images that are lewd, explicit, or, otherwise, offensive. In step 340, the computing device may convert each image, of the subset of images, to a vector graphic and / or a line drawing. In particular, the computing device may create computer-generated images that are composed of geometric shapes, lines, and points using mathematical formulas. The vector graphics and / or line drawings may be used for a variety of applications, including print and digital media. The computing devicePCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 may use one or more applications, such as Potrace, to convert each of the subset of images to a vector graphic.

[0040] In step 345, the computing device may cause the vector graphic associated with each image to be displayed. The vector graphics may be displayed as a grid or in any other suitable format. The display may allow the user to save selections, for example, by clicking a like or heart icon. Saved selections may be stored locally in the user device, for example, in a cache or in cookies. Saved selections may be recovered from the local storage, for example, if the user ends the session without selecting a final design. The user may resume the previous session from where they previously left off.

[0041] In step 350, the computing device may receive a selection of a first vector graphic. In response to receiving the selection, the computing device may cause the first vector graphic to be displayed on a product, such as drinkware, in step 355. Displaying the first vector graphic may comprise displaying the first vector graphic as an overlay on the product. In step 360, the computing device may receive confirmation of the first vector graphic on the product. In response to receiving confirmation of the first vector graphic on the product, the computing device may cause a product to be generated with the first vector graphic, in step 365. Producing the product with the first vector graphic may include etching (e.g., laser etching) the first vector graphic on to the product. Additionally or alternatively, producing the product with the first vector graphic may comprise printing the first vector graphic on the product. Printing the first vector graphic on the product may use techniques similar to inkjet printing to produce color images on the product.

[0042] The process described above may be used to create images using generative artificial intelligence and converting those images to vector graphics. FIGS. 4A-4D show an example of an interface for creating images using generative artificial intelligence in accordance with one or more aspects of the disclosure.

[0043] FIG. 4A shows a web browser 405. Web browser 405 may display a webpages associated with customizing drinkware 402. The webpage displayed via web browser 405 may display a customization bar 410. Customization bar 410 may include a plurality of options for customizing drinkware 402. The plurality of options may comprise selecting a color, selecting a design for the front of dnnkware 402, and / or selecting a design for the back of drinkware 402. The webpage may provide several options for the design. For example, the user may be able to upload a design, enterPCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 text to be printed on drinkware 402, select an image from a gallery, or enter a monogram to be printed on drinkware 402. In some instances, the user may choose to have generative artificial intelligence create a design for drinkware 402 by selecting Al design button 415.

[0044] When a user selects the Al design button 415, the webpage displayed via web browser 405 may be updated, similar to the display shown in FIG. 4B. In this regard, selecting Al design button 415 may cause window 420 to be displayed via web brow'ser 405. Window 420 may include a text prompt 425. While text prompt 425 is shown in FIG. 4B, it will be appreciated that the prompt may be any type of prompt, including a microphone for a user to enter a description of the design or an option to upload an image of the design. In this regard, text prompt 425 may allow the user to enter a description of the design the user would like to have on drinkware 402. As shown in FIG. 4B, the user has entered “dog with surfboard” in text prompt 425.

[0045] After clicking “OK,” the input may be analyzed to ensure that the input complies with one or more policies, as discussed above. When the input complies with the one or more policies, the input may be modified to include one or more conditions. As noted above, the one or more conditions may be designed to produce results that may be more suitable for conversion to vector graphics. The modified input may be provided to one or more generative artificial intelligence models, which generate one or more images based on the modified input. The one or more images may be analyzed to ensure that the one or more images comply with the one or more policies. A subset of images may be selected from the one or more images. The selected subset of images may be converted to vector graphics and displayed via the user device, as shown in FIG. 4C.

[0046] In this regard, FIG. 4C shows web browser 405 displaying a results window 430. Results window 430 includes a subset of four images. However, it will be appreciated that the subset of images displayed in results window 430 may include more, or fewer, images. Results window 430 may allow a user to select a design that the user would like to be reproduced on drinkware 402. Although not shown in FIG. 4C, each result may be selected as a preferred image, for example, by clicking a “Like” button, a heart-shaped button, or any similar type of button. The preferred image may be stored locally on the user device, for example, in a cache or cookies. A user may select a design from results window 430, for example, by clicking, or double-clicking, on a design.PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136

[0047] Once the user selects a design, the design may be displayed on drinkware 402, as shown in FIG. 4D. Web browser 405 may be updated to include drinkware 402, with the overlain image, and “Buy” button 435. “Buy” button 435 allows the user to confirm that the user would like to purchase drinkware 402 with the selected image reproduced thereon. By clicking the “Buy” button 435 and entering payment information, a version of drinkware 402 may be produced that includes the selected image. As discussed above, the image may be reproduced on drinkware 402 via a variety of techniques, including laser etching and printing.

[0048] The techniques described above improve generative artificial intelligence models’ ability to output vector graphic images or images more suitable for conversion to vector graphics. In particular, the present disclosure leverages prompt engineering by modifying the input to a generative artificial intelligence model to produce vector graphic images or images more suitable for conversion to vector graphics. Moreover, the present disclosure uses a variety of conversion techniques to ensure that the images produced by generative artificial intelligence models are capable of being reproduced as vector graphics.

[0049] One or more features discussed herein may be embodied in computer-usable or readable data and / or computer-executable instructions, such as in one or more program modules, executed by one or more computers or other devices as described herein. Program modules may comprise routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types when executed by a processor in a computer or other device. The modules may be written in a source code programming language that is subsequently compiled for execution, or may be written in a scripting language such as (but not limited to) Python, Perl, or any equivalent thereof. The computer executable instructions may be stored on a computer readable medium such as a hard disk, optical disk, removable storage media, solid-state memory, RAM, and the like. The functionality of the program modules may be combined or distributed as desired. In addition, the functionality may be embodied in whole or in part in firmware or hardware equivalents such as integrated circuits, field programmable gate arrays (FPGA), and the like. Particular data structures may be used to more effectively implement one or more features discussed herein, and such data structures are contemplated within the scope of computer executable instructions and computer-PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 usable data described herein. Various features described herein may be embodied as a method, a computing device, a system, and / or a computer program product.

[0050] Although the present disclosure has been described in terms of various examples, many additional modifications and variations would be apparent to those skilled in the art. In particular, any of the various processes described above may be performed in alternative sequences and / or in parallel (on different computing devices) in order to achieve similar results in a manner that is more appropriate to the requirements of a specific application. It is therefore to be understood that the present disclosure may be practiced otherwise than specifically described without departing from the scope and spirit of the present disclosure. Although examples are described above, features and / or steps of those examples may be combined, divided, omitted, rearranged, revised, and / or augmented in any desired manner. Thus, the present disclosure should be considered in all respects as illustrative and not restrictive. Accordingly, the scope of the disclosure should be determined not by the examples, but by the appended claims and their equivalents.

Claims

PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136CLAIMS1. A method comprising: receiving, by a computing device via a prompt of an interface, an input; generating, based on the input and using generative artificial intelligence, a plurality of images; analyzing, using one or more machine learning models, the plurality of images to determine whether the plurality of images complies with one or more policies; based on a determination that a subset of images, of the plurality of images, complies with the one or more policies, convert each image, of the subset of images, to a vector graphic; causing at least one vector graphic associated with an image, of the subset of images, to be displayed; causing the at least one vector graphic, overlain on a product, to be displayed; and in response to receiving confirmation of the at least one vector graphic on the product, generating the product with the at least one vector graphic.

2. The method of claim 1, wherein the receiving the input comprises receiving a written description of a user’s request.

3. The method of claim 1, wherein the receiving the input comprises receiving an upload of an image or a photo.

4. The method of claim 1, further comprising: analyzing, prior to generating the plurality of images, the input to determine whether the input complies with the one or more policies, wherein the generating the plurality of images is based on a determination that the input complies with the one or more policies.

5. The method of claim 1, further comprising: modifying, prior to the generating the plurality of images, the input.

6. The method of claim 5, wherein modifying the input comprises adding one or more negative conditions to the input.

7. The method of claim 1, wherein generating the plurality of images further comprises:PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 transmitting, to the generative artificial intelligence and via one or more application programming interfaces, the input; and receiving, from the generative artificial intelligence, the plurality of images.

8. The method of claim 1, wherein the generating the product with the at least one vector graphic comprises at least one of: laser etching the at least one vector graphic on drinkware; or printing the at least one vector graphic on drinkware.

9. The method of claim 8, wherein the at least one vector graphic comprises a color image.

10. The method of claim 1, further comprising: receiving, by the computing device via the prompt of the interface, a second input; analyzing the second input to determine whether the input complies with the one or more policies; and based on a determination that the second input does not comply with the one or more policies, causing an error message to be displayed.

11. The method of claim 10, wherein the error message comprises one or more instructions to revise the second input.

12. The method of claim 1, further comprising: causing a plurality of vector graphics to be displayed concurrently with the at least one vector graphic; and receiving a selection of the at least one vector graphic, wherein the at least one vector graphic is displayed in response to the selection.

13. The method of claim 1, wherein the one or more machine learning models comprise one or more: large language models (LLM(s)); convolutional neural networks (CNN); recurrent neural networks; recursive neural networks; long short-term memory (LSTM);PCT / US25 / 49394 03 October 2025 (03.10.2025)008117.11136 gated recurrent units (GRU); unsupervised pre-trained networks; space invariant artificial neural networks; generative adversarial networks (GAN); consistent adversarial networks (CAN); cyclic generative adversarial networks (C-GAN); deep convolutional GANs (DC-GAN);GAN interpolation (GAN-INT);GAN-CLS; or a cychc-CAN (C-CAN).

14. A computing device comprising: one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the computing device to perform the method of any one of claims 1-13.

15. A non-transitory computer-readable medium storing instructions that, when executed, cause a computing device to perform the method of any one of claims 1-13.

Images