Big data recommendation image platform based on AI intelligent interaction

Abstract

The invention discloses a big data recommendation image platform based on AI intelligent interaction, and relates to the technical field of image design. The system specifically comprises a data acquisition module which performs three-dimensional data acquisition through an image or a camera and constructs a personal three-dimensional data model of a user; the data acquisition module can modify the body shape data of the user and simulate the expected body shape of the user; and the condition analysis module is used for selecting the corresponding hair style, makeup and clothes according to the keywords selected and input by the user, and simulating the image of the user. The model of the user is quickly modified through the artificial intelligence interaction module, it is ensured that the model conforms to the aesthetic appreciation of the user, and the use effect of the platform is improved; meanwhile, the platform rapidly establishes a three-dimensional data model of the user through the data acquisition module, and cooperates with the condition analysis module to rapidly generate an adaptive image model according to the requirements of the user.

Description

Technical Field

[0001] The present invention relates to the field of image design technology, and in particular to a big data image recommendation platform based on AI intelligent interaction. Background Art

[0002] Image design, also known as image consultant, mostly uses professional diagnostic tools to take body color as the basic feature and comprehensive factors such as face and body shape, temperament and social role.

[0003] After searching, the Chinese patent application No. 201910761049.1 discloses a personalized image design platform, including a personalized design platform, the personalized setting platform includes an input system, a processing system, an element storage system and an output display system; the input system includes an image input module and a text input module, the image input module is used to input the user's head and full-body image, and the text input module is used to input the user's occupation, gender, age and other characteristic information; the processing system includes an image analysis module, a text analysis module and a comparative analysis module, the image analysis module includes a portrait analysis module and an image processing module, and the portrait analysis module is used to analyze the facial features of the portrait, The facial image is processed to obtain the length-to-width ratio of the face, the position of the mandible, the position of the cheekbones, and the curvature of the face to calculate the shape and size parameters of the face, and finally obtain the information of the face shape and facial features. The image processing module is used to analyze the contour of the full-body image to obtain the user's figure, body shape and height information. The text analysis module analyzes the information data input by the user and extracts valid information from the input data. The comparison analysis module is used to search and compare the parameter information of the image processed in the image processing analysis module and the key data information extracted by the text analysis module with the hairstyle, face shape, body shape and clothing in the element storage system to obtain the hairstyle and clothing that matches the user.

[0004] The personalized image design platform in the above patent has the following shortcomings: the design platform is convenient for people to refer to, which can reduce people's unnecessary expenses when designing their personal image, and can facilitate people to choose a personalized image that suits them; however, in actual use, it is not convenient for users to quickly adjust the design image of the platform according to their own aesthetic characteristics and actual needs, thereby reducing the overall use effect of the personal image design platform. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a big data image recommendation platform based on AI intelligent interaction.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A big data image recommendation platform based on AI intelligent interaction, including: A data acquisition module, which collects three-dimensional data through images or cameras to construct a three-dimensional data model of the user; and the data acquisition module can modify the user's body shape data to simulate the user's desired body shape; A condition analysis module, which selects corresponding hairstyle, makeup and clothing according to the keywords selected and input by the user to simulate the image of the user; A database module, wherein the database module includes a human body three-dimensional database and an image database, wherein the human body three-dimensional database includes a face sub-database and a body shape sub-database, and the image database includes a hairstyle sub-database, a makeup sub-database and a clothing sub-database; An image recommendation and display module, which constructs a basic data image of the user based on the personal three-dimensional data model constructed by the data acquisition module and the user image simulated by the condition analysis module, and can make adjustments to the basic data image according to the needs of the user; The artificial intelligence interaction module performs human-computer interaction with the user, controls the data collection module and the condition analysis module, and controls the image recommendation display module to perform personal image recommendation, display and adjustment.

[0007] Preferably: the data acquisition module constructs a personal three-dimensional basic model through the user's front, back, left, right headshots, half-body shots and full-body shots; and the data acquisition module can also use a camera to take all-round surround shots of the user to construct the personal three-dimensional basic model, and use the camera to collect real-time data on the user's skin condition; the human body three-dimensional database provides data support for the data acquisition module.

[0008] Furthermore: the data acquisition module fine-tunes the personal three-dimensional basic model through the manually input data of head circumference, neck circumference, chest circumference, waist circumference, hip circumference, head length, arm length, leg length, foot length, height and weight to construct a personal body simulation model; and the data acquisition module collects detailed data on the user's head through a camera to construct a personal face simulation model of the user; the personal body simulation model and the personal face simulation model are fused with the user's skin data to construct a personal three-dimensional data model.

[0009] As a preferred embodiment of the present invention: the conditional analysis module determines the keywords selected and input by the user for his or her own image through the artificial intelligence interaction module, and the artificial intelligence interaction module provides big data support for the conditional analysis module. With the support of big data, the conditional analysis module selects the user's hairstyle, makeup and clothing according to the user's keywords.

[0010] On the basis of the above scheme: the image database provides data support for the conditional analysis module, the hairstyle sub-database includes bangs, sideburns, back ponytail, and hair color, the makeup sub-database includes eyebrows, eyes, nose, mouth, gills, cheeks, forehead, and neck; the clothing sub-database includes headwear, earrings, jewelry, tops, bottoms, and shoes.

[0011] Preferably: the image recommendation and display module integrates the personal three-dimensional data model with the hairstyle, makeup and clothing models simulated by the condition analysis module to form the user's basic data image and display it; the user directly retrieves data from the database module through the image recommendation and display module and adjusts the positions that need to be modified.

[0012] Furthermore: the artificial intelligence interaction module constructs the underlying aesthetic logic through big data, including the matching logic of clothing, the matching logic of makeup and clothing, the matching logic of hairstyle and makeup, the matching logic of clothing and body shape, the matching logic of hairstyle and body shape, and the matching logic of hair color and skin color.

[0013] Based on the above scheme: the underlying aesthetic logic of the artificial intelligence interaction module is improved through big data review, and the improvement steps include: S1: Preliminary learning of big data: Based on big data, learn a large number of public modeling pictures, and deepen the aesthetic logic based on the underlying aesthetic logic; S2: construct a basic data image for testing; automatically generate multiple groups of personal three-dimensional data models through the data acquisition module; each group of personal three-dimensional data models is set to multiple, and each personal three-dimensional data model automatically matches multiple keywords; the condition analysis module automatically matches the shape for each personal three-dimensional data model to generate the corresponding basic data image; S3: Aesthetic logic optimization: multiple professional modeling designers analyze, evaluate, and modify the basic data image to obtain the optimized data image, and upload the optimized data image to the database for learning by the artificial intelligence interaction module to further optimize the underlying aesthetic logic; S4: Repeat S2 and S3 until the created basic data image meets the usage requirements.

[0014] The beneficial effects of the present invention are: This big data image recommendation platform based on AI intelligent interaction can quickly modify the user's appearance through the artificial intelligence interaction module to ensure that the appearance conforms to the user's own aesthetic taste and improve the use effect of the platform; at the same time, this platform can quickly establish the user's own three-dimensional data model through the data acquisition module, and cooperate with the condition analysis module to quickly generate an adaptive image model according to the user's requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the overall modular structure of the present invention. DETAILED DESCRIPTION

[0016] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.

[0017] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0018] A big data image recommendation platform based on AI intelligent interaction, such as Figure 1 As shown, it includes: data collection module, condition analysis module, database module, image recommendation display module and artificial intelligence interaction module.

[0019] The data acquisition module collects three-dimensional data through images or cameras to build a three-dimensional data model of the user; and the data acquisition module can modify the user's body shape data to simulate the user's desired body shape; The data collection module constructs a personal three-dimensional basic model through the user's front, back, left, right and right headshots, half-length photos and full-length photos; the constructed personal three-dimensional basic model is as close to the user as possible, ensuring that the image after styling design is close to the user.

[0020] The data acquisition module can also use the camera to take all-around surround shots of the user, collect real-time data of the user, and build a personal three-dimensional basic model in real time to ensure the convenience of building the personal three-dimensional basic model.

[0021] The data collection module collects real-time data on the user's skin condition through a camera to ensure the accuracy of the user's skin color, luster and other data, thereby ensuring that the image of the styling design is consistent with the user's skin.

[0022] The data acquisition module fine-tunes the personal three-dimensional basic model through the manually input data of head circumference, neck circumference, chest circumference, waist circumference, hip circumference, head length, arm length, leg length, foot length, height and weight to construct a personal body simulation model.

[0023] The data acquisition module collects detailed data on the user's head through a camera, such as eyes, eyebrows, mouth, nose, etc., to build a personal facial simulation model of the user; the personal body simulation model and the personal facial simulation model are combined with the user's skin data to build a personal three-dimensional data model.

[0024] The weight, waist circumference and hip circumference in the personal body simulation model can be adjusted to a relatively large extent to generate a control model of the personal body simulation model, which is convenient for users to plan the optimization of their own body shape according to their own situation.

[0025] The conditional analysis module selects the corresponding hairstyle, makeup and clothing based on the keywords selected and input by the user to simulate the user's image.

[0026] The conditional analysis module determines the keywords selected and input by the user for his or her own image through the artificial intelligence interaction module, and selects the hairstyle, makeup and clothing corresponding to the user based on the data of the personal three-dimensional data model.

[0027] The artificial intelligence interaction module provides big data support for the conditional analysis module. With the support of big data, the conditional analysis module adaptively optimizes the hairstyle, makeup and clothing according to the personal three-dimensional data model to construct a corresponding image.

[0028] The database module includes a human body three-dimensional database and an image database. The human body three-dimensional database includes a face sub-database and a body shape sub-database. The image database includes a hairstyle sub-database, a makeup sub-database and a clothing sub-database. The human body three-dimensional database provides data support for the data acquisition module. The body shape sub-database includes height, weight, head circumference, neck circumference, chest circumference, waist circumference, hip circumference, head length, arm length, leg length, and foot length; the facial sub-database includes eye size, eye distance, mouth size, face size, cheekbone parameters, forehead parameters, nose size, ear size, etc.

[0029] The image database provides data support for the conditional analysis module. The hairstyle sub-database includes bangs, sideburns, back ponytail, and hair color. The makeup sub-database includes eyebrows, eyes, nose, mouth, gills, cheeks, forehead, and neck. The clothing sub-database includes headwear, earrings, jewelry, tops, bottoms, and shoes.

[0030] The image recommendation and display module constructs the user's basic data image based on the personal three-dimensional data model constructed by the data collection module and the user image simulated by the condition analysis module, and can make adjustments to the basic data image according to the user's needs; The image recommendation and display module integrates the personal three-dimensional data model with the hairstyle, makeup and clothing models simulated by the condition analysis module to form the user's basic data image and display it, which is determined by the user based on his or her own aesthetic taste.

[0031] If the user is not satisfied with the basic data image, the image recommendation display module can directly retrieve data from the database module and adjust the position that needs to be modified so that the modified data image meets the user's own aesthetic needs.

[0032] The artificial intelligence interaction module conducts human-computer interaction with the user, and controls the data collection module and the condition analysis module to collect data from the user, and process and analyze the user's requirements for their own image modeling, while controlling the image recommendation display module to perform personal image recommendation, display and adjustment.

[0033] The artificial intelligence interaction module constructs the underlying aesthetic logic through big data, including the matching logic of clothing, the matching logic of makeup and clothing, the matching logic of hairstyle and makeup, the matching logic of clothing and body shape, the matching logic of hairstyle and body shape, and the matching logic of hair color and skin color.

[0034] After the underlying aesthetic logic of the AI ​​interaction module is constructed, it is improved through big data review. The improvement steps include: S1: Preliminary learning of big data: Based on big data, learn a large number of public modeling pictures, and deepen the aesthetic logic based on the underlying aesthetic logic; analyze the public modeling pictures and split them into personal body simulation models, skin data, personal face simulation models, makeup data, clothing data and hairstyle data; connect the relationship between the above data and the underlying aesthetic logic, so that the artificial intelligence interaction module can initially deepen the connection between the underlying aesthetic logic and the actual modeling design; S2: Construct a basic data image for testing; automatically generate multiple groups of personal 3D data models through the data acquisition module, the multiple groups of personal 3D data models are based on height, weight, and thinness, and contain multiple facial data combinations; each group of personal 3D data models is set to multiple, each personal 3D data model automatically matches multiple keywords, and the keywords of each personal 3D data model are different; the condition analysis module automatically matches the shape for each personal 3D data model to generate the corresponding basic data image; S3: Aesthetic logic optimization: multiple professional modeling designers analyze, evaluate, and modify the basic data image to obtain the optimized data image, and upload the optimized data image to the database for learning by the artificial intelligence interaction module to further optimize the underlying aesthetic logic; S4: Repeat S2 and S3 until the created basic data image meets the usage requirements.

[0035] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A big data image recommendation platform based on AI intelligent interaction, characterized in that: include: A data acquisition module, which collects three-dimensional data through images or cameras to construct a three-dimensional data model of the user; and the data acquisition module can modify the user's body shape data to simulate the user's desired body shape; A condition analysis module, which selects corresponding hairstyle, makeup and clothing according to the keywords selected and input by the user to simulate the image of the user; A database module, wherein the database module includes a human body three-dimensional database and an image database, wherein the human body three-dimensional database includes a face sub-database and a body shape sub-database, and the image database includes a hairstyle sub-database, a makeup sub-database and a clothing sub-database; An image recommendation and display module, which constructs a basic data image of the user based on the personal three-dimensional data model constructed by the data acquisition module and the user image simulated by the condition analysis module, and can make adjustments to the basic data image according to the needs of the user; The artificial intelligence interaction module performs human-computer interaction with the user, controls the data collection module and the condition analysis module, and controls the image recommendation display module to perform personal image recommendation, display and adjustment.

2. According to claim 1, a big data image recommendation platform based on AI intelligent interaction is characterized by: The data acquisition module constructs a personal three-dimensional basic model through the user's front, back, left, right and right headshots, half-body photos and full-body photos; and the data acquisition module can also use a camera to take all-round surround photos of the user to construct a personal three-dimensional basic model, and use the camera to collect real-time data on the user's skin condition; the human body three-dimensional database provides data support for the data acquisition module.

3. According to claim 1, a big data image recommendation platform based on AI intelligent interaction is characterized by: The data acquisition module fine-tunes the personal three-dimensional basic model through the manually input data of head circumference, neck circumference, chest circumference, waist circumference, hip circumference, head length, arm length, leg length, foot length, height and weight to construct a personal body simulation model; and the data acquisition module collects detailed data on the user's head through a camera to construct the user's personal face simulation model; the personal body simulation model and the personal face simulation model are combined with the user's skin data to construct a personal three-dimensional data model.

4. According to claim 1, a big data image recommendation platform based on AI intelligent interaction is characterized by: The conditional analysis module determines the keywords selected and input by the user for his / her own image through the artificial intelligence interaction module. The artificial intelligence interaction module provides big data support for the conditional analysis module. With the support of big data, the conditional analysis module selects the user's hairstyle, makeup and clothing based on the user's keywords.

5. According to claim 1, a big data image recommendation platform based on AI intelligent interaction is characterized by: The image database provides data support for the conditional analysis module; the hairstyle sub-database includes bangs, sideburns, back ponytail, and hair color; the makeup sub-database includes eyebrows, eyes, nose, mouth, gills, cheeks, forehead, and neck; the clothing sub-database includes headwear, earrings, jewelry, tops, bottoms, and shoes.

6. The big data image recommendation platform based on AI intelligent interaction according to claim 1, characterized in that: The image recommendation and display module combines the personal three-dimensional data model with the hairstyle, makeup and clothing models simulated by the condition analysis module to form the basic data image of the user and display it; The user can directly retrieve data from the database module through the image recommendation display module and adjust the positions that need to be modified.

7. The big data image recommendation platform based on AI intelligent interaction according to claim 1 is characterized by: The artificial intelligence interaction module constructs the underlying aesthetic logic through big data, including the matching logic of clothing, the matching logic of makeup and clothing, the matching logic of hairstyle and makeup, the matching logic of clothing and body shape, the matching logic of hairstyle and body shape, and the matching logic of hair color and skin color.

8. The big data image recommendation platform based on AI intelligent interaction according to claim 7 is characterized by: The underlying aesthetic logic of the artificial intelligence interaction module is improved through big data review, and the improvement steps include: S1: Preliminary learning of big data: Based on big data, learn a large number of public modeling pictures, and deepen the aesthetic logic based on the underlying aesthetic logic; S2: construct a basic data image for testing; automatically generate multiple groups of personal three-dimensional data models through the data acquisition module; each group of personal three-dimensional data models is set to multiple, and each personal three-dimensional data model automatically matches multiple keywords; the condition analysis module automatically matches the shape for each personal three-dimensional data model to generate the corresponding basic data image; S3: Aesthetic logic optimization: multiple professional modeling designers analyze, evaluate, and modify the basic data image to obtain the optimized data image, and upload the optimized data image to the database for learning by the artificial intelligence interaction module to further optimize the underlying aesthetic logic; S4: Repeat S2 and S3 until the created basic data image meets the usage requirements.

Images