Intelligent generation system and method of AI sticker

Abstract

The invention relates to the technical field of AI intelligent imaging, in particular to an intelligent generation system and method for an AI sticker, and the system comprises a parameter setting module, an image collection module, an AI processing module, and a sticker output module. The parameter setting module is used for setting AI stylization parameters and camera parameters; the image acquisition module is used for shooting images based on the camera parameters; the AI processing module is used for carrying out AI processing on the image based on the AI stylization parameter, generating a sticker photo and obtaining a contour frame of the sticker photo; and the sticker output module is used for uploading the sticker photo to a cloud end and printing and outputting the sticker photo based on the contour frame. The portrait AI is stylized in a photographing mode, the stylized photo is printed into the sticker, the whole process is automatic, the operation process is greatly simplified, the efficiency is improved, and meanwhile the quality and the individuation degree of the sticker are improved.

Description

Technical Field

[0001] The present invention relates to the field of AI intelligent imaging technology, and in particular to an intelligent generation system and method for AI stickers. Background Art

[0002] With the development of AI technology, portrait style conversion has become a hot research direction, which can realize the conversion of portrait photos into various artistic styles, such as cartoonization, sketching, oil painting, etc. However, most of these converted images are only used for online sharing rather than physical applications. On the other hand, various sticker printing devices on the market can print customized stickers according to user needs. However, these devices usually require users to manually import, adjust and crop images, which is complicated and inefficient.

[0003] Defects and shortcomings of the existing technology:

[0004] High operation complexity: For existing sticker printing equipment, users need to perform multiple steps to complete the printing and cutting of stickers, including importing images, adjusting image size and position, setting printing parameters, etc. The operation process is complicated and not conducive to user use.

[0005] Low efficiency: Most existing sticker printing devices require users to manually import and crop images, which not only increases the user's operating burden, but also greatly reduces production efficiency.

[0006] Limited stylization processing capabilities: Although there are some software on the market that can perform stylization processing, most of these software can only convert images into several preset styles and often require users to make manual adjustments, which cannot meet users' needs for diverse and personalized styles.

[0007] Separation of AI stylization and sticker production: In existing technologies, AI stylization processing and sticker production are usually two independent processes. After completing the stylization processing, users need to import the image into the sticker printing device for printing and cutting. The operation process is cumbersome and the user experience is poor.

[0008] The quality of stylized images is not high: Existing AI stylization technology may not be able to guarantee the quality of processed images when processing complex images or high-quality images, resulting in poor printed stickers.

[0009] In view of the above shortcomings, the present invention proposes an intelligent generation system and method of AI stickers. The portrait AI is stylized by taking photos, and the stylized photos are printed into stickers. The whole process is automated, which greatly simplifies the operation process and improves efficiency, while also improving the quality and personalization of stickers. Summary of the invention

[0010] The purpose of the present invention is to provide an intelligent generation system and method for AI stickers, which can stylize portrait AI by taking photos, and print the stylized photos into stickers. The whole process is automated, which greatly simplifies the operation process and improves efficiency, while also improving the quality and personalization of stickers.

[0011] To achieve the above object, the present invention provides the following solutions:

[0012] An intelligent generation system for AI stickers, comprising:

[0013] Parameter setting module, image acquisition module, AI processing module, sticker output module;

[0014] The parameter setting module is used to set AI stylization parameters and camera parameters;

[0015] The image acquisition module is used to capture images based on the camera parameters;

[0016] The AI ​​processing module is used to perform AI processing on the image based on the AI ​​stylization parameters, generate a sticker photo and obtain a contour frame of the sticker photo;

[0017] The sticker output module is used to upload the sticker photo to the cloud and print out the sticker photo based on the outline frame.

[0018] Optionally, the parameter setting module includes: an AI stylization parameter setting unit and a camera parameter setting unit;

[0019] The AI ​​stylization parameter setting unit is used for the user to select an AI style and set AI stylization parameters, and save them to the cloud;

[0020] The camera parameter setting unit is used for the user to set the camera parameters and connect the display device, and then test shoot based on the camera parameters to check whether the camera parameters meet the preset requirements. If they meet, the parameters are saved; if not, the parameters are adjusted.

[0021] Optionally, the image acquisition module includes: an image acquisition unit and an image uploading unit;

[0022] The image acquisition unit is used for users to take photos that meet personal requirements;

[0023] The image uploading unit is used to upload photos to the cloud for storage and processing.

[0024] Optionally, the AI ​​processing module includes: a style selection unit, an AI processing unit, a sticker adjustment unit, and a contour frame marking unit;

[0025] The style selection unit is used for the user to select an AI style;

[0026] The AI ​​processing unit is used to render the image based on the AI ​​stylization parameters and the AI ​​style selected by the user to obtain an AI stylized sticker photo;

[0027] The sticker adjustment unit is used to obtain the position information parameters of the AI ​​stylized sticker photo and perform a personalized adjustment operation on the AI ​​stylized sticker photo to obtain a final sticker photo;

[0028] The outline frame marking unit is used to mark the outline frame of the final sticker photo based on the position information parameters.

[0029] Optionally, obtaining the location information parameters of the AI ​​stylized sticker photo includes:

[0030] Convert the AI ​​stylized sticker photo into a Sprite object and add it to the parent container Container, determine whether the attribute parameters are legal, if not, return conversion failure, if legal, parse the public parameters of the AI ​​stylized sticker photo and store them in the object GlobalStyle;

[0031] Set the center point, draw the text object based on the center point and the parsed common parameters, and locate the absolute position information of the text object relative to the parent container Container, that is, the position information parameters of the AI ​​stylized sticker photo.

[0032] Optionally, the sticker output module includes: a cloud uploading unit, a printing selection unit, and a cutting and printing unit;

[0033] The cloud uploading unit is used to upload the final sticker photo to the cloud and generate a unique QR code for the user to scan and obtain the final sticker photo;

[0034] The printing selection unit is used for the user to select the sticker photos and the number of copies to be printed;

[0035] The cutting and printing unit is used to perform cutting and printing based on the outline frame of the selected sticker photo to output a finished sticker.

[0036] To further achieve the above objectives, the present invention also provides an intelligent generation method of AI stickers, comprising:

[0037] Set AI stylization parameters and camera parameters;

[0038] Taking an image based on the camera parameters, performing AI processing on the image based on the AI ​​stylization parameters, and obtaining a sticker photo and a contour frame of the sticker photo;

[0039] The sticker photo is uploaded to the cloud and the sticker photo is printed out based on the outline frame to obtain a finished sticker.

[0040] Optionally, performing AI processing on the image based on the AI ​​stylization parameter includes:

[0041] Select AI style;

[0042] Rendering the image based on the AI ​​stylization parameters and the AI ​​style to obtain an AI stylized sticker photo;

[0043] Obtain the position information parameters of the AI ​​stylized sticker photo and perform personalized adjustment operations on the AI ​​stylized sticker photo to obtain a final sticker photo, and mark the outline frame of the final sticker photo based on the position information parameters.

[0044] Optionally, obtaining the location information parameters of the AI ​​stylized sticker photo includes:

[0045] Convert the AI ​​stylized sticker photo into a Sprite object and add it to the parent container Container, determine whether the attribute parameters are legal, if not, return conversion failure, if legal, parse the public parameters of the AI ​​stylized sticker photo and store them in the object GlobalStyle;

[0046] Set the center point, draw the text object based on the center point and the parsed common parameters, and locate the absolute position information of the text object relative to the parent container Container, that is, the position information parameters of the AI ​​stylized sticker photo.

[0047] The beneficial effects of the present invention are:

[0048] Automated operation: The entire process of the present invention, from taking pictures, AI stylization processing, printing stickers to automatic cutting, is all automated. Users only need to perform simple operations to complete it, which greatly simplifies the operation process and improves the user experience.

[0049] Diverse artistic styles: Through AI stylization processing, users can choose a variety of artistic styles for stylization, and can customize adjustments according to their needs, meeting users' needs for personalized stickers.

[0050] High efficiency: Since the whole process is automated, the manual operation time is greatly reduced and the work efficiency is improved. Especially in the occasion where a large number of stickers need to be produced, the present invention can significantly improve the production efficiency.

[0051] High sticker quality: The printing device can automatically adjust the printing parameters according to the size and quality of the image to ensure that the printed stickers have excellent effects and meet the user's high requirements for sticker quality.

[0052] Eliminate manual cutting steps: After printing, the sticker is sent to the cutting machine through an automatic transmission device for cutting. The cutting machine automatically cuts according to the edge of the image, eliminating the manual cutting steps and reducing the work intensity.

[0053] Cloud processing, low device requirements: The AI ​​stylization process is completed in the cloud, which has low performance requirements on user devices, allowing more users to use this technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 This is a schematic diagram of the structure of an intelligent generation system for AI stickers according to an embodiment of the present invention;

[0056] Figure 2 A method for using and a workflow diagram of an intelligent generation system for AI stickers according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] This embodiment provides an intelligent generation system for AI stickers, such as Figure 1 As shown, it includes: parameter setting module, image acquisition module, AI processing module, and sticker output module;

[0060] Parameter setting module, used to set AI stylization parameters and camera parameters;

[0061] An image acquisition module, used for capturing images based on camera parameters;

[0062] An AI processing module, used to perform AI processing on the image based on AI stylization parameters, generate a sticker photo and obtain a contour frame of the sticker photo;

[0063] The sticker output module is used to upload sticker photos to the cloud and print out the sticker photos based on the outline frame.

[0064] Further, the parameter setting module includes: an AI stylization parameter setting unit and a camera parameter setting unit;

[0065] AI stylization parameter setting unit, used for users to select AI style and set AI stylization parameters, and save them to the cloud;

[0066] The camera parameter setting unit is used by the user to set the camera parameters and connect the display device. After taking a test shot based on the camera parameters, the camera parameters are checked to see if they meet the preset requirements. If they meet, they are saved; if not, they are adjusted.

[0067] The image acquisition module includes: an image acquisition unit and an image uploading unit;

[0068] An image acquisition unit, used for users to take photos that meet personal requirements;

[0069] Image upload unit, used to upload photos to the cloud for storage and processing.

[0070] Further, the AI ​​processing module includes: a style selection unit, an AI processing unit, a sticker adjustment unit, and a contour frame marking unit;

[0071] Style selection unit, used by users to select AI style;

[0072] An AI processing unit, used to render the image based on AI stylization parameters and the AI ​​style selected by the user, and obtain an AI stylized sticker photo;

[0073] A sticker adjustment unit, used to obtain position information parameters of the AI ​​stylized sticker photo and perform personalized adjustment operations on the AI ​​stylized sticker photo to obtain a final sticker photo;

[0074] The outline frame marking unit is used to mark the outline frame of the final sticker photo based on the position information parameters.

[0075] Specifically, the location information parameters for obtaining AI stylized sticker photos include:

[0076] Convert the AI ​​stylized sticker photo into a Sprite object and add it to the parent container Container. Check whether the attribute parameters are legal. If not, return conversion failure. If legal, parse the public parameters of the AI ​​stylized sticker photo and store them in the object GlobalStyle.

[0077] Set the center point, draw the text object based on the center point and the parsed common parameters, and locate the absolute position information of the text object relative to the parent container Container, that is, the position information parameters of the AI ​​stylized sticker photo.

[0078] Furthermore, the sticker output module includes: a cloud upload unit, a print selection unit, and a cutting and printing unit;

[0079] The upload cloud unit is used to upload the final sticker photo to the cloud and generate a unique QR code for users to scan to obtain the final sticker photo;

[0080] Print selection unit, used for users to select sticker photos and number of copies to be printed;

[0081] The cutting and printing unit is used to perform cutting and printing based on the outline frame of the selected sticker photo and output the finished sticker.

[0082] To further optimize the technical solution, this embodiment also provides an intelligent generation method of AI stickers, including:

[0083] Set AI stylization parameters and camera parameters;

[0084] Take an image based on the camera parameters, perform AI processing on the image based on the AI ​​stylization parameters, and obtain the sticker photo and the outline frame of the sticker photo;

[0085] Upload the sticker photo to the cloud and print out the sticker photo based on the outline frame to obtain the finished sticker.

[0086] Furthermore, performing AI processing on the image based on the AI ​​stylization parameters includes:

[0087] Select AI style;

[0088] Render the image based on AI stylization parameters and AI style to obtain AI stylized sticker photos;

[0089] The location information parameters of the AI ​​stylized sticker photo are obtained and personalized adjustments are performed on the AI ​​stylized sticker photo to obtain the final sticker photo, and the outline frame of the final sticker photo is marked based on the location information parameters.

[0090] Specifically, the AI ​​stylization in this embodiment can make the character into an AI image and then make it into a sticker, or the character can be made into a sticker, but in the form of a real person + sticker.

[0091] Specifically, the location information parameters for obtaining AI stylized sticker photos include:

[0092] Convert the AI ​​stylized sticker photo into a Sprite object and add it to the parent container Container. Check whether the attribute parameters are legal. If not, return conversion failure. If legal, parse the public parameters of the AI ​​stylized sticker photo and store them in the object GlobalStyle.

[0093] Set the center point, draw the text object based on the center point and the parsed common parameters, and locate the absolute position information of the text object relative to the parent container Container, that is, the position information parameters of the AI ​​stylized sticker photo.

[0094] Combine the following Figure 2 The usage method and workflow of the intelligent generation system of AI stickers proposed in this embodiment are described in detail. First, use a camera to take a portrait photo. This step can be performed at any location. You only need to ensure that the light in the shooting environment is sufficient and the portrait is clear. After the shooting is completed, the portrait photo is converted into an image with a specific artistic style through the AI ​​stylization processing algorithm in the cloud. The AI ​​stylization processing algorithm can adopt a neural network style transfer technology based on deep learning, support the conversion of multiple artistic styles, and can be customized according to user needs. After the stylization processing is completed, the software will automatically send the image to the connected sticker printing device for printing. The printing device can automatically adjust the printing parameters according to the size and quality of the image to ensure that the printed sticker has excellent effect. After printing is completed, the sticker is sent to the cutting machine for cutting through an automatic transmission device, and the cutting machine automatically cuts according to the edge of the image.

[0095] The system uses the user's interactive input as the AI ​​style parameter input, then collects the user's real-time photo data, embeds it into a variety of photo templates for typesetting, and then processes the image algorithm. Then the user's interaction is used again as a secondary preview template selection, photo frame selection, signature customization, etc., and finally local synthesis, printing, uploading, and cloud AR synthesis are performed.

[0096] The system sends the thumbnail of the photo to the cloud, and then the cloud calls the AI ​​stylization parameters through the API, and then returns the image to the local after AI stylization processing.

[0097] The use method and workflow of an AI sticker intelligent generation system in this embodiment specifically include the following steps:

[0098] 1. Set AI sticker style parameters:

[0099] Set the parameters of AI stylized stickers in the cloud, including style type, color, style features, etc.

[0100] Users can select these parameters through the application interface to ensure that the generated stickers meet their personal needs.

[0101] 2. Set up camera and printer:

[0102] Adjust the parameters of the camera so that it can be connected to the tablet and ensure that the portrait photos taken are clear and meet the requirements.

[0103] Connect the tablet to the printer via a USB data cable to ensure stable data transmission.

[0104] 3. Check camera connection and photo quality:

[0105] Check whether the camera connection is normal and ensure that the quality of the photos taken meets the requirements.

[0106] If you have problems connecting to the camera or the picture quality is not satisfactory, go back and reset it.

[0107] 4. Start taking photos:

[0108] The user clicks the start photo button to take a portrait photo.

[0109] The photos are automatically uploaded to the tablet and displayed on the app interface.

[0110] 5. Upload photos to the cloud:

[0111] Photos are uploaded to the cloud for processing.

[0112] Check whether the upload is successful. If it fails, return an upload failure notification.

[0113] 6. Select AI style:

[0114] Users browse on the screen and select their favorite AI style.

[0115] The user confirms the selected AI style for further processing.

[0116] 7. Image rendering:

[0117] The cloud performs stylized processing on photos based on the set AI parameters.

[0118] Once processed, the resulting AI-stylized sticker photo is returned to the tablet and displayed.

[0119] Check if rendering is successful. If it fails, return the processing failure information.

[0120] 8. Convert AI stylized stickers to Sprite objects:

[0121] Convert the AI ​​stylized stickers into Sprite objects and add them to the parent container Container.

[0122] Checks whether the conversion is appropriate and returns conversion failure information if it fails.

[0123] 9. Parse parameters and draw stickers:

[0124] Parse the common parameters and store them into the GlobalStyle object.

[0125] Determine whether fontType in the public parameters is a number. If not, get the Chinese or English configuration name.

[0126] Set the common parameters to center alignment and set the center point to (0,0);

[0127] Draw the text object according to specific parameters and common parameters.

[0128] According to the attribute parameters, locate the absolute position information of the text object relative to the parent container Container and add it to the parent container Container.

[0129] 10. Add adjustment operation events to the parent container:

[0130] Users can move, scale, rotate, delete, adjust transparency, and perform other operations on the generated sticker objects to achieve personalized adjustments.

[0131] 11. Mark the photo outline frame;

[0132] Use the AI ​​function to select the outline frame of the obtained AI stylized sticker photo, convert the outline frame into a lettering path, and send it to the lettering machine.

[0133] 12. Select the number of copies to print:

[0134] The user selects the number of copies to print in the application.

[0135] The user confirms the selected number of copies to print.

[0136] 13. Start printing and cutting:

[0137] The printer receives the data and starts printing the stylized sticker.

[0138] Check whether printing is successful. If it fails, return the printing failure information.

[0139] 14. Send the sticker position information parameters to the engraving machine:

[0140] After printing is successful, the system will send the sticker location information parameters and engraving path to the engraving machine.

[0141] 15. Determine whether the sending is successful:

[0142] Check whether the sticker position information parameters and engraving path are successfully sent to the engraving machine.

[0143] If the sending fails, a sending failure notification is returned.

[0144] If the sending is successful, the engraving machine starts engraving along the edge of the sticker.

[0145] 16. Engraving machine for Yanbian sticker engraving:

[0146] After receiving the signal, the engraving machine starts engraving along the edge of the sticker to ensure that the shape and edges of the sticker are accurate.

[0147] 17. Get the finished sticker:

[0148] The user takes the finished sticker from the engraving machine.

[0149] At the same time, users can scan the QR code through the application to obtain electronic version of AI-style sticker photos for easy sharing and saving.

[0150] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. An intelligent generation system for AI stickers, It is characterized in that include: Parameter setting module, image acquisition module, AI processing module, sticker output module; The parameter setting module is used to set AI stylization parameters and camera parameters; The image acquisition module is used to capture images based on the camera parameters; The AI ​​processing module is used to perform AI processing on the image based on the AI ​​stylization parameters, generate a sticker photo and obtain a contour frame of the sticker photo; The sticker output module is used to upload the sticker photo to the cloud and print out the sticker photo based on the outline frame.

2. The intelligent generation system of AI stickers according to claim 1, It is characterized in that The parameter setting module includes: an AI stylization parameter setting unit and a camera parameter setting unit; The AI ​​stylization parameter setting unit is used for the user to select an AI style and set AI stylization parameters, and save them to the cloud; The camera parameter setting unit is used for the user to set the camera parameters and connect the display device, and then test shoot based on the camera parameters to check whether the camera parameters meet the preset requirements. If they meet, the parameters are saved; if not, the parameters are adjusted.

3. The intelligent generation system of AI stickers according to claim 1, It is characterized in that The image acquisition module includes: an image acquisition unit and an image uploading unit; The image acquisition unit is used for users to take photos that meet personal requirements; The image uploading unit is used to upload photos to the cloud for storage and processing.

4. The intelligent generation system of AI stickers according to claim 1, It is characterized in that The AI ​​processing module includes: a style selection unit, an AI processing unit, a sticker adjustment unit, and a contour frame marking unit; The style selection unit is used for the user to select an AI style; The AI ​​processing unit is used to render the image based on the AI ​​stylization parameters and the AI ​​style selected by the user to obtain an AI stylized sticker photo; The sticker adjustment unit is used to obtain the position information parameters of the AI ​​stylized sticker photo and perform a personalized adjustment operation on the AI ​​stylized sticker photo to obtain a final sticker photo; The outline frame marking unit is used to mark the outline frame of the final sticker photo based on the position information parameters.

5. The intelligent generation system of AI stickers according to claim 4, It is characterized in that The parameters for obtaining the location information of the AI ​​stylized sticker photo include: Convert the AI ​​stylized sticker photo into a Sprite object and add it to the parent container Container, determine whether the attribute parameters are legal, if not, return conversion failure, if legal, parse the public parameters of the AI ​​stylized sticker photo and store them in the object GlobalStyle; Set the center point, draw the text object based on the center point and the parsed common parameters, and locate the absolute position information of the text object relative to the parent container Container, that is, the position information parameters of the AI ​​stylized sticker photo.

6. The intelligent generation system of AI stickers according to claim 4, It is characterized in that The sticker output module includes: a cloud uploading unit, a printing selection unit, and a cutting and printing unit; The cloud uploading unit is used to upload the final sticker photo to the cloud and generate a unique QR code for the user to scan and obtain the final sticker photo; The printing selection unit is used for the user to select the sticker photos and the number of copies to be printed; The cutting and printing unit is used to perform cutting and printing based on the outline frame of the selected sticker photo to output a finished sticker.

7. An intelligent generation method of AI stickers, used to implement the intelligent generation system of AI stickers as described in any one of claims 1 to 6, It is characterized in that include: Set AI stylization parameters and camera parameters; Taking an image based on the camera parameters, performing AI processing on the image based on the AI ​​stylization parameters, and obtaining a sticker photo and a contour frame of the sticker photo; The sticker photo is uploaded to the cloud and the sticker photo is printed out based on the outline frame to obtain a finished sticker.

8. The intelligent generation method of AI stickers according to claim 7, It is characterized in that Performing AI processing on the image based on the AI ​​stylization parameter includes: Select AI style; Rendering the image based on the AI ​​stylization parameters and the AI ​​style to obtain an AI stylized sticker photo; Obtain the position information parameters of the AI ​​stylized sticker photo and perform personalized adjustment operations on the AI ​​stylized sticker photo to obtain a final sticker photo, and mark the outline frame of the final sticker photo based on the position information parameters.

9. The intelligent generation method of AI stickers according to claim 8, It is characterized in that The parameters for obtaining the location information of the AI ​​stylized sticker photo include: Convert the AI ​​stylized sticker photo into a Sprite object and add it to the parent container Container, determine whether the attribute parameters are legal, if not, return conversion failure, if legal, parse the public parameters of the AI ​​stylized sticker photo and store them in the object GlobalStyle; Set the center point, draw the text object based on the center point and the parsed common parameters, and locate the absolute position information of the text object relative to the parent container Container, that is, the position information parameters of the AI ​​stylized sticker photo.

Images