A method for drawing a body shape vector view of a real size

Abstract

This invention addresses the need for personalized clothing to depict human body morphological features by providing a method for creating a vector view of the human body based on measured dimensions. The method involves using image recognition technology to obtain the width, thickness, and height dimensions of various circumferences (including chest, waist, and hip circumference) from a photograph of the human body. The method calculates the ratio of the major and minor axes and the ratio of the height of each rounded circumference section to determine its shape and position. Based on the principles of anthropometric charts for clothing, and combined with the circumference and height values, a vector view of the human body morphology is created. This view breaks away from the existing method of using abstract graphics for clothing anthropometric charts, providing an effective means of expressing human body dimensions and morphological features within the scope of clothing design. It offers a reference for pattern design and sewing process design of customized clothing, provides a basis for the realistic expression of human body shape, and ultimately lays the foundation for human body morphology identification.

Description

Technical Field

[0001] This invention relates to the field of clothing and human body drawing technology, specifically a method for drawing a vector view of human body shape based on measured dimensions. Background Technology

[0002] In the apparel industry, human body dimensions and physique are crucial bases for garment design and engineering. Human circumference and morphological characteristics serve as important references for garment styling. Currently, apparel companies often use a single human body measurement as the basis for garment styling, neglecting human morphological characteristics. Garment pattern designs for different body types use a consistent proportion for design and production; for example, the front and back panel proportions of garment patterns ignore human circumference and morphological characteristics. This is one of the main reasons why custom-made garments often fail to fit properly.

[0003] To address the aforementioned issues, this invention presents a vector view that incorporates both human body circumference and morphological characteristics into clothing design.

[0004] This view breaks away from the traditional impressionistic drawing method and is drawn using the method of measuring the human body circumference and reconstructing the human body circumference cross-sectional shape. It vectorizes the human body shape characteristics, enriches the way of expressing the real human body shape characteristics, can meet the needs of personalized clothing production for human body shape characteristics, and also helps people understand the correspondence between two-dimensional human body size and three-dimensional clothing shape. Summary of the Invention

[0005] This invention addresses the need for personalized clothing to depict human body features by providing a method for drawing a vector view of human form based on measured dimensions.

[0006] The drawing method specifically includes the following steps:

[0007] Obtain the width, thickness, and height dimensions of each circumference from the human photograph. These circumferences include chest, waist, and hip circumference. Calculate the ratio of the major and minor axes and the ratio of the height of each circumference cross-section after rounding, thereby determining the cross-sectional shape and position of each circumference. Based on the principles of human anatomy diagrams for clothing and human circumference and height values, draw a vector view of the human form. The specific steps are as follows:

[0008] Step S1: Determine the requirements for the pre-drawn human body view for clothing. The view drawing data includes the dimensions of the human body height control parts and the dimensions of the human body circumference control parts, as well as the circumference shape representation values. The view consists of three views: front, side, and back.

[0009] Step S2: Measure or obtain the circumference series dimensions of each group of body types Y, A, B, and C for the specified size;

[0010] Step S3: Collect frontal and side-view photos of the subject according to the set human posture acquisition requirements;

[0011] Step S4: Process the collected human body photos to form an analyzable human body contour image. Use intelligent image algorithms to identify and extract the width and thickness of each circumference section of the human body and the height value of the human body, and calculate the ratio of the major axis to the minor axis and the ratio of the height of each circumference of the human body.

[0012] Step S5: Based on the requirements of the pre-drawn human body vector view in step S1 and the implementation results of steps S2 and S4, draw the front, side, and back three views of the human body shape vector.

[0013] As a preferred option: the dimensions of the human body height control parts are proportional values, specifically including the ratio of neck circumference, chest circumference, waist circumference, hip circumference to height; the dimensions of the human body circumference control parts include chest circumference, waist circumference, and hip circumference; the circumference shape characterization value is the ratio of the major axis to the minor axis of the rounded circumference cross section.

[0014] As a preferred method, the drawing method is applicable to people of normal body types, specifically the Y, A, B, and C body types covered by GB / T 1335.1-2008 and GB / T1335.2-2008.

[0015] Preferably, the requirements for capturing human posture are as follows: For a frontal view, the subject should stand naturally upright with shoulders naturally open, eyes looking straight ahead, arms extended and aligned with the center line of the side of the body, maintaining a certain distance from the body, and feet level with the shoulders; for a side view, the subject should adjust to have arms hanging naturally at their sides, not obscuring the lowest point of the waist, and feet together. The subject should be fully visible in the camera, with the camera positioned at half the subject's height, horizontally above the ground.

[0016] Preferably, the analyzable human contour image is obtained by separating the human body from the background in the photo using the MODNet segmentation algorithm and by image noise processing, which includes opening and closing operations and image edge detection.

[0017] Preferably, the steps of the intelligent image algorithm to identify and extract the width and thickness of each circumference section of the human body and the height of the human body are as follows:

[0018] Step i. Establish a human body reference coordinate system based on the analyzable human body contour image;

[0019] Step ii. Based on the human contour model and conventional human skeleton features, set human feature points and target feature points on the image that are related to the size to be acquired;

[0020] Step iii. Based on the feature points set in step ii, extract the coordinate values ​​of human body feature points, and calculate the target feature points according to the human body structure and image mapping relationship;

[0021] Step iv. Based on the results of step iii, extract the major and minor axis values ​​and height dimensions of each circumference, including bust, waist, hip, neck height, bust height, waist height, and hip height.

[0022] Preferably, the image setting of human body feature points includes identifying existing feature points and inferred feature points. Existing feature points are the curve undulation points of the human body model or skeletal feature points, while inferred feature points are hidden feature points inferred based on existing feature points and the principles of human body structure.

[0023] Preferably, the feature points of the same circumference in the frontal and side profile human body contour models are mathematically mapped to each other, and the mathematical mapping relationship is as follows:

[0024] The mapping relationship between the front and side chest circumference points can be expressed by equation (1): ;

[0025] In the formula, It is the vertical distance between the horizontal line where the side waistline point and the front neckline point are located;

[0026] The vertical distance between the horizontal lines containing the front bust point and the front neck point;

[0027] The vertical distance between the horizontal lines containing the frontal anterior neck point and the frontal mid-hip point;

[0028] The vertical distance between the horizontal lines containing the anterior neck point and the mid-hip point on the side;

[0029] The mapping relationship between the front and side waist circumference points can be expressed by equation (2):

[0030] ;

[0031] In the formula, The vertical distance between the horizontal line containing the front waistline and the front mid-hip point;

[0032] The vertical distance between the horizontal line containing the side waistline point and the side mid-hip point;

[0033] The vertical distance between the horizontal lines containing the frontal anterior neck point and the frontal mid-hip point;

[0034] The vertical distance between the horizontal lines containing the anterior neck point and the mid-hip point on the side;

[0035] The mapping relationship between the front and side hip circumference points can be expressed by equation (3):

[0036] ;

[0037] In the formula, The vertical distance between the horizontal line containing the front hip circumference point and the front mid-hip point;

[0038] It is the vertical distance between the horizontal line containing the side hip circumference point and the side mid-hip point;

[0039] The vertical distance between the horizontal lines containing the frontal anterior neck point and the frontal mid-hip point;

[0040] It is the vertical distance between the horizontal lines containing the anterior neck point and the mid-hip point on the side.

[0041] This invention uses image recognition technology to obtain the width, thickness, and height dimensions of the chest, waist, and hip circumference from acquired human photographs. It calculates the ratio of the major axis to the minor axis and the ratio of the height of each rounded circumference section, thereby determining the cross-sectional shape and position of each rounded circumference section. Based on the principles of human body structure diagrams for clothing and the circumference and height values ​​of four body types (Y, A, B, and C), a vector view of the human body shape is drawn.

[0042] The advantages of implementing and applying this invention are as follows: This invention provides a method for drawing vector views of human body shape based on measured dimensions. It uses the straight-line distance of each measured height dimension and the ratio of the major and minor axes of the rounded circumference section to determine the position of the circumference section and the shape representing the circumference. Combining this with the principles of anthropometric drawing for clothing, it draws three vector views: front, back, and side. This breaks through the existing method of drawing anthropometric drawings for clothing, which are all impressionistic graphics. It provides an effective expressive medium for incorporating human body circumference dimensions and morphological characteristics into the scope of clothing design, providing a reference for pattern design and sewing process design of customized clothing, and also providing a basis for the realistic expression of human body shape, thus laying the foundation for human body shape identification. This invention, by determining the shape and position of the cross-section of each rounded circumference and combining it with the principles of anthropometric drawing for clothing, draws vector views. Compared to the drawing of impressionistic graphics, this method is highly practical and can be drawn by anyone with a little knowledge of anthropometric drawing for clothing. Attached Figure Description

[0043] Figure 1 shows the process of drawing a vector view of the human body shape based on actual measured dimensions;

[0044] Figure 2 shows the existing distribution of human feature points;

[0045] Figure 3 shows the distribution of human feature points required for the embodiment;

[0046] Figure 4 is a schematic diagram for determining the position of the frontal bust point;

[0047] Figure 5 shows the mapping relationship between the front and side views of the chest circumference.

[0048] Figure 6 is a schematic diagram for determining the location of the side bust measurement point;

[0049] Figure 7 shows the mapping relationship between the front and side waist circumference points;

[0050] Figure 8 shows the mapping relationship between the front and side hip circumference points;

[0051] Figure 9 shows the vector views of the Y-shaped body: (a) front view, (b) rear view, and (c) side view.

[0052] Figure 10 shows the vector view of body shape A, (a) is the front view, (b) is the back view, and (c) is the side view;

[0053] Figure 11 shows the vector view of body shape B, (a) is the front view, (b) is the back view, and (c) is the side view;

[0054] Figure 12 shows the vector view of the C-shaped body, (a) is the front view, (b) is the back view, and (c) is the side view. Detailed Implementation

[0055] The drawing method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Those skilled in the art will be able to implement the present invention based on these descriptions. Furthermore, the embodiments of the present invention described below are generally only some, not all, of the embodiments of the present invention. Therefore, all other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort should fall within the scope of protection of the present invention.

[0056] Figure 1 shows the process of drawing a vector view of human body shape based on actual measured dimensions in this example. It mainly includes three modules: determining the requirements for drawing a human body view for clothing, obtaining the circumference and height values ​​of the human body, and obtaining the cross-sectional axis ratio and height ratio.

[0057] Module 1: Determine the requirements for pre-drawing the human body view for clothing.

[0058] The view drawing data includes the dimensions of the human body height control parts and the dimensions of the human body circumference control parts, as well as numerical values ​​representing the circumference shape. The view consists of three views: front, side, and back. The position of the circumference section and the representation of the circumference shape are determined by the straight-line distance of each height dimension and the ratio of the major and minor axes of the circular cross-section. In this example, the human body height dimensions include height, neck circumference, chest circumference, waist circumference, and hip circumference, and the human body circumference dimensions include neck circumference, chest circumference, waist circumference, and hip circumference.

[0059] Module 2: Obtain body circumference and height values

[0060] Referring to the four body types Y, A, B, and C specified in GB / T 1335.1-2008, this implementation extracts the circumference series dimensions and height values ​​for each group of the four body types under the set size, as shown in Table 1.

[0061]

[0062] Module 3: Obtain the ratio of the major axis to the minor axis and the height value of the circular cross-section.

[0063] Human body photography was conducted. Two subjects, each 170 cm tall with slightly different body types, were selected for this embodiment. Both body types conformed to the standard body type defined in GB / T 1335.1-2008. Frontal and side-view photographs of the subjects were taken according to the photography requirements. For the frontal photograph, the subject stood naturally upright with shoulders naturally open, eyes looking straight ahead, arms extended and aligned with the center line of the side of the body, maintaining a certain distance from the body, and feet level with the shoulders. For the side-view photograph, the subject adjusted their position so that arms hung naturally at their sides, not obscuring the lowest point of the waist, and feet together. The subject was fully visible in the photograph, with the camera positioned at half the subject's height, horizontally above the ground.

[0064] Photo preprocessing creates analyzable images, including separating the human body from the background and image noise reduction. The MODNet segmentation algorithm is used to separate the human body from the background in photos of two subjects, resulting in a highly targeted and accurate human body segmentation model. Image noise reduction includes opening and closing operations, as well as image edge detection. Opening operations eliminate objects other than the human body identified in the image, smooth shape boundaries without changing the area. Closing operations fill small holes within objects in the image and connect broken contour lines. Finally, the Canny operator is used for image edge detection to obtain a human body contour model.

[0065] Identify the human body contour in an image and establish a human body reference coordinate system.

[0066] Based on the human contour model and conventional human skeletal features, human feature points and target feature points related to the acquired dimensions are set on the image, as shown in Figures 2 and 3. The scatter plot in Figure 2 represents existing human feature points obtained from the identification of human contour and skeletal position lines. Figure 3 shows the distribution of existing feature points to be extracted in this implementation. The feature points to be extracted in this implementation include the side waistline point, shoulder point, upper end of the humerus, lower end of the humerus, frontal and side anterior neck point, side hipline point, and frontal and side mid-hip point, etc., and their names are shown in Table 2. The target feature points in this embodiment include the frontal and side chestline points, frontal and side waistline points, and frontal and side hipline points, used to calculate the major and minor axis values ​​of the chest, waist, and hip cross-sections.

[0067] Table 2. Names of Human Feature Points in Examples

[0068]

[0069] The OpenPose algorithm is used to extract the coordinates of human body feature points. Target feature points, including front and side chest circumference points, front and side waist circumference points, and front and side hip circumference points, are calculated based on human body structure and image mapping relationships. The positions of other target feature points are deduced based on the principles of human body structure and the mathematical mapping relationship between the front and side images.

[0070] Based on the principles of human anatomy, the front and side chest circumference points are calculated as follows:

[0071] 1) The chest curve is covered by clothing, and the front chest point cannot be identified from the photo. However, the horizontal line where the midpoint of the humerus is located in a normal body type is approximately the horizontal line of the chest point. The intersection of this horizontal line and the line connecting the shoulder point and the side waist point is the front chest point, as shown in Figure 4.

[0072] 2) Based on the mathematical mapping relationship between the two sets of images, the side chest circumference can be determined from the front chest circumference point and other human feature points identified from the images. Formula (1) is its data conversion formula, and the determination principle is as follows: Figure 5 As shown.

[0073] 3) For a standard body type, the chest measurement point falls on the perpendicular line connecting the front neck point and the mid-hip point. This allows determination of the lateral chest measurement point. Figure 6 As shown.

[0074] Determine the front waistline and side hipline based on the mathematical mapping relationship between the two sets of images:

[0075] 1) Use formula (2) to calculate and determine the position of the front waistline point. The principle is as follows: Figure 7 As shown.

[0076] 2) The position of the side hip circumference point is calculated and determined using formula (3). The determination principle is shown in Figure 8.

[0077] The major and minor axes and height dimensions of each circumference are calculated based on the coordinates of characteristic points such as the front and side chest circumference points, front and side waist circumference points, front and side hip circumference points, and neck circumference points. These dimensions include, but are not limited to, chest circumference, waist circumference, hip circumference, neck circumference height, chest circumference height, waist circumference height, and hip circumference height. The ratio of the major and minor axes and the ratio of the height of the human body circumference cross-section are then calculated.

[0078] Two sets of data were obtained by collecting human photos of the two subjects and performing the above operations, as shown in Table 3.

[0079] Table 3. Aspect Ratio and Height Ratio of Male Human Body Circumference Section

[0080]

[0081] Based on the data listed in Tables 1 and 3, draw vector views of the four body types: Y, A, B, and C, as shown below. Figures 9-12 As shown in Figures 9-12, these are vector views of the front, back, and side views of four body types: Y, A, B, and C. The views clearly express the differences in morphological characteristics of the different body types, demonstrating the significance and application of this invention.

Claims

1. A method for drawing a vector view of human body shape based on measured dimensions, characterized in that: Obtain the width, thickness, and height dimensions of various measurements from the anatomy photograph, including bust, waist, and hip circumference. Calculate the ratio of the major axis to the minor axis and the ratio of the height of each circumference section after smoothing, thereby determining the cross-sectional shape and dimensions of each circumference. Based on the principles of human anatomy diagrams for clothing and human circumference and height values, a vector view of the human form is drawn. The steps are as follows: Step S1: Determine the requirements for the pre-drawn human body view for clothing. The view drawing data includes the dimensions of the human body height control parts and the dimensions of the human body circumference control parts, as well as the circumference shape representation values. The view consists of three views: front, side, and back. Step S2: Measure or obtain the circumference series dimensions of each group of body types Y, A, B, and C in the specified size; Step S3: Collect frontal and side-view photos of the subject according to the set collection requirements; Step S4: Process the collected human body photos to form an analyzable human body contour image. Use intelligent image algorithms to identify and extract the width and thickness of each circumference section of the human body and the height value of the human body, and calculate the ratio of the major axis to the minor axis and the ratio of the height of each circumference of the human body. Step S5: Based on the requirements of the pre-drawn human body vector view in step S1 and the implementation results of steps S2 and S4, draw the front, side, and back three views of the human body shape vector. The steps of the intelligent image algorithm to identify and extract the width and thickness of various circumference sections of the human body and the height of the human body are as follows: Step i. Establish a human body reference coordinate system based on the analyzable human body contour image; Step ii. Based on the human contour model and conventional human skeleton features, set human feature points and target feature points on the image that are related to the size to be acquired; Step iii. Based on the feature points set in step ii, extract the coordinate values ​​of human body feature points, and calculate the target feature points according to the human body structure and image mapping relationship; Step iv. Based on the results of step iii, extract the major and minor axis values ​​and height dimensions for each circumference, including bust, waist, hips, neck height, etc. High bust, high waist, high hips; The image setting of human body feature points includes identifying existing feature points and inferred feature points. Existing feature points are the curve undulation points of the human body model or skeletal feature points, while inferred feature points are hidden feature points inferred based on existing feature points and the principles of human body structure. The feature points of the same circumference in the front and side profile human body contour models are mathematically mapped to each other, and the mathematical mapping relationship is as follows: The mapping relationship between the front and side chest circumference points can be expressed by equation (1): ； In the formula, The vertical distance between the side waistline point and the horizontal line connecting the front neckline point; The vertical distance between the horizontal lines containing the front bust point and the front neck point; The vertical distance between the horizontal lines containing the frontal anterior neck point and the frontal mid-hip point; The vertical distance between the horizontal lines containing the anterior neck point and the mid-hip point on the side; The mapping relationship between the front and side waist circumference points can be expressed by equation (2): ； In the formula, The vertical distance between the horizontal line connecting the waistline point and the mid-hip point at the front; The vertical distance between the side waistline point and the horizontal line containing the side mid-hip point; The mapping relationship between the front and side hip circumference points can be expressed by equation (3): ； In the formula, The vertical distance between the horizontal line containing the front hip circumference point and the front mid-hip point; It is the vertical distance between the horizontal line where the side hip circumference point and the side mid-hip point are located.

2. The method for drawing a vector view of human body shape based on measured dimensions according to claim 1, characterized in that: The dimensions of body height control points are proportional values, specifically including the ratios of neck circumference, chest circumference, waist circumference, and hip circumference to height; the dimensions of body circumference control points include chest circumference, ... Waist circumference and hip circumference; the circumference shape representation value is the ratio of the major axis to the minor axis of the rounded circumference cross section.

3. The method for drawing a vector view of human body shape by measuring dimensions according to claim 1, characterized in that: The analyzable human contour image is obtained by separating the human body from the background in the photo using the MODNet segmentation algorithm and by image noise processing, which includes opening and closing operations and image edge detection.

Images