Virtual fitting mirror system

Abstract

A virtual fitting mirror system disclosed by the present invention comprises a base, the upper side of the base is fixedly connected with a mirror frame, the interior of the mirror frame is fixedly connected with one-way perspective glass and a display screen, the one-way perspective glass and the display screen respectively occupy two halves in the mirror frame, and the back side of the one-way perspective glass is fixedly connected with a mounting box. A camera is fixedly connected into the mounting box, the front side of the camera is a reflecting mirror surface, the back side of the camera is a perspective mirror surface, and the camera can shoot pictures on the front side through the back side of the one-way perspective glass. A user can conveniently check different colors and patterns of the same clothes style, the user does not need to carry out multiple times of fitting, convenience and rapidness are achieved, and the satisfaction degree of the fitting effect can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of clothing matching, and in particular to a virtual fitting mirror system. Background Art

[0002] Clothing matching mainly refers to the coordination of styles and colors to achieve a decent and generous effect overall. As for the classification of personal clothing styles and the dressing methods of various styles, how to quickly and accurately give the style type results that conform to the individual's own rules, so that ordinary people's dressing life becomes scientific and regular, this problem is difficult to solve. In this era of self-promotion and individuality, more people pay more attention to personal clothing matching according to their own circumstances and preferences to show different personalities and selves.

[0003] Currently, when users are trying on clothes, since the same style of clothes may have different colors and patterns, in order to get satisfaction from the fitting effect, users need to try on different colors and patterns of the same style of clothes multiple times, which is rather cumbersome and troublesome. Therefore, innovation is needed to solve the above problem. Summary of the invention

[0004] The present invention provides a virtual fitting mirror system to solve the problems raised in the above background technology.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A virtual fitting mirror system comprises a base, wherein a mirror frame is fixedly connected to the upper side of the base, a one-way perspective glass and a display screen are fixedly connected to the interior of the mirror frame, the one-way perspective glass and the display screen occupy two halves of the mirror frame respectively, a mounting box is fixedly connected to the back side of the one-way perspective glass, a camera is fixedly connected to the mounting box, the front side of the camera is a reflective mirror surface, the back side of the camera is a perspective mirror surface, and the camera can take pictures of the front side through the back side of the one-way perspective glass;

[0007] It also includes a control system, which includes a character outline interception module, a clothing type database, and a virtual fitting module.

[0008] Preferably, the character outline capture module is used to process the image of the user in front of the one-way perspective glass taken by the camera to obtain the character outline image of the user.

[0009] Preferably, the method for constructing the character outline interception module is:

[0010] S1: Obtain a large number of specimen images containing human figures;

[0011] S2: marking the figure contour area in the specimen image containing the figure of the person, to obtain a marked specimen image of the figure of the person;

[0012] S3: Based on the labeled human body specimen image and annotation information, a human contour extraction module is obtained through detection model training.

[0013] Preferably, the detection model is a model obtained by a Mask R-CNN deep learning network.

[0014] Preferably, the specimen image containing the human figure is subjected to pixel binarization processing.

[0015] Preferably, during training, the input information is divided into a prediction set and a training set. After the training, the validity of the model is verified in the test set. If the verification fails, the S one-way transparent glass is used to add more labeled data; if the verification passes, the training ends.

[0016] Preferably, if the average AP value of the test set is <0.95, the verification fails; if the average AP value is ≥0.95, the verification passes.

[0017] Preferably, the clothing type database is used to store a large number of clothing model drawings.

[0018] Preferably, the virtual fitting module is used to display the character outline image obtained by the character outline capture module on a display screen, and identify the clothing style on the user and the color and pattern of the style, then retrieve other colors and patterns of the clothing style in the clothing type database, and finally replace the color and pattern of the clothing style on the user with different colors and patterns.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention allows users to conveniently view different colors and patterns of the same clothing style without the need for users to try on multiple times, which is not only convenient and fast, but also helps to improve satisfaction with fitting results.

[0021] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. The specific implementation of the present invention is given in detail by the following embodiments and their accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0023] Figure 1 A schematic diagram of the three-dimensional structure of a virtual fitting mirror system proposed by the present invention;

[0024] Figure 2 A schematic diagram of a front cross-sectional structure of a virtual fitting mirror system proposed by the present invention;

[0025] Figure 3 This is a schematic diagram of the control system structure proposed by the present invention.

[0026] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0027] 1. Mirror frame; 2. One-way perspective glass; 3. Camera; 4. Installation box; 5. Base; 6. Display screen. DETAILED DESCRIPTION

[0028] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention. The present invention is described in more detail by way of example with reference to the accompanying drawings in the following paragraphs. It should be noted that the accompanying drawings are all in a very simplified form and are not in precise proportions, and are only used to facilitate and clearly assist in explaining the purpose of the embodiments of the present invention.

[0029] It should be noted that when a component is referred to as being "fixed to" another component, it may be directly on the other component or there may also be a component centered. When a component is considered to be "connected to" another component, it may be directly connected to the other component or there may also be a component centered. When a component is considered to be "set on" another component, it may be directly set on the other component or there may also be a component centered. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0031] See also Figures 1 to 3In an embodiment of the present invention, a virtual fitting mirror system includes a base 5, a mirror frame 1 is fixedly connected to the upper side of the base 5, a one-way perspective glass 2 and a display screen 6 are fixedly connected to the inside of the mirror frame 1, the one-way perspective glass 2 and the display screen 6 occupy two halves of the mirror frame 1 respectively, a mounting box 4 is fixedly connected to the back side of the one-way perspective glass 2, a camera 3 is fixedly connected to the mounting box 4, the front side of the camera 3 is a reflective mirror surface, the back side of the camera 3 is a perspective mirror surface, and the camera 3 can take a picture of the front side through the back side of the one-way perspective glass 2;

[0032] It also includes a control system, which includes a character outline capture module, a clothing type database, and a virtual fitting module.

[0033] The character outline capture module is used to process the image of the user in front of the one-way perspective glass 2 taken by the camera 3 to obtain the character outline image of the user.

[0034] The clothing type database is used to store a large number of clothing model drawings;

[0035] The virtual fitting module is used to display the character outline image obtained by the character outline interception module on the display screen 6, and identify the clothing style of the user and the color and pattern of the style, then retrieve other colors and patterns of the clothing style in the clothing type database, and finally replace the color and pattern of the clothing style on the user with different colors and patterns.

[0036] The construction method of the character outline interception module is:

[0037] S1: obtaining a large number of specimen images containing human figures, and each specimen image containing human figures undergoes pixel binarization processing to make the image clearer;

[0038] S2: marking the figure contour area in the specimen image containing the figure of the person, and obtaining a marked specimen image of the figure of the person;

[0039] S3: Based on the labeled human body specimen image and annotation information, the human contour extraction module is obtained through detection model training, wherein the detection model is a Mask R-CNN deep learning network model; during training, the input information is divided into a prediction set and a training set. After the training, the effectiveness of the model is verified in the test set. If the verification fails, it is transferred to S2 to add more labeled data; if the verification passes, the training ends. If the average AP value of the test set is <0.95, the verification fails; if the average AP value is ≥0.95, the verification passes.

[0040] The working principle of the present invention is:

[0041] When in use, the user can stand in front of 2, and then 3 can take a full-body photo of the user's body. Then the character outline capture module is used to process the picture of the user in front of the one-way perspective glass 2 taken by camera 3 to obtain the user's character outline image. The virtual fitting module then displays the character outline image obtained by the character outline capture module on the display screen 6, and identifies the style of clothes on the user and the color and pattern of the style, and then retrieves other colors and patterns of the clothes style in the clothes type database, and finally replaces the color and pattern of the user's clothes style with different colors and patterns and displays them on the display screen 6. The user can conveniently view different colors and patterns of the same clothes style, thereby improving the satisfaction with the fitting effect.

[0042] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any ordinary technician in the industry can smoothly implement the present invention as shown in the drawings and described above. However, any equivalent changes, modifications and evolutions made by technicians familiar with the profession without departing from the scope of the technical solution of the present invention using the technical content disclosed above are all equivalent embodiments of the present invention. At the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the technical solution of the present invention.

Claims

1. A virtual fitting mirror system, comprising a base (5), characterized in that: The upper side of the base (5) is fixedly connected to a mirror frame (1), the interior of the mirror frame (1) is fixedly connected to a one-way perspective glass (2) and a display screen (6), the one-way perspective glass (2) and the display screen (6) occupy two halves of the mirror frame (1) respectively, the back side of the one-way perspective glass (2) is fixedly connected to a mounting box (4), the mounting box (4) is fixedly connected to a camera (3), the front side of the camera (3) is a reflective mirror surface, the back side of the camera (3) is a perspective mirror surface, and the camera (3) can take a picture of the front side through the back side of the one-way perspective glass (2); It also includes a control system, which includes a character outline interception module, a clothing type database, and a virtual fitting module.

2. A virtual fitting mirror system according to claim 1, characterized in that: The character outline capture module is used to process a picture of a user in front of the one-way perspective glass (2) taken by a camera (3) to obtain a character outline image of the user.

3. A virtual fitting mirror system according to claim 2, characterized in that: The construction method of the character outline interception module is: S1: Obtain a large number of specimen images containing human figures; S2: marking the figure contour area in the specimen image containing the figure of the person, to obtain a marked specimen image of the figure of the person; S3: Based on the labeled human body specimen image and annotation information, a human contour extraction module is obtained through detection model training.

4. A virtual fitting mirror system according to claim 3, characterized in that: The detection model is a model obtained by the Mask R-CNN deep learning network.

5. A virtual fitting mirror system according to claim 4, characterized in that: The specimen image containing the human figure is subjected to pixel binarization processing.

6. A virtual fitting mirror system according to claim 5, characterized in that: During training, the input information is divided into a prediction set and a training set. After the training is completed, the validity of the model is verified in the test set. If the verification fails, it is transferred to S2 to add more labeled data; If the verification passes, the training ends.

7. A virtual fitting mirror system according to claim 6, characterized in that: If the average AP value of the test set is <0.95, the verification fails; if the average AP value is ≥0.95, the verification passes.

8. A virtual fitting mirror system according to claim 7, characterized in that: The clothing type database is used to store a large number of clothing model drawings.

9. A virtual fitting mirror system according to claim 8, characterized in that: The virtual fitting module is used to display the human figure contour image obtained by the human figure contour interception module on a display screen (6), and identify the clothing style of the user and the color and pattern of the style, then retrieve other colors and patterns of the clothing style from the clothing type database, and finally replace the color and pattern of the clothing style of the user with different colors and patterns.

Images