# US 2024/0202931 A1 · Measuring method and system for body-shaped data

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0202931 A1
- Document ID: US20240202931A1
- Status: Published application
- Kind: A1
- Application number: 18/080,847
- Filed: 2022-12-14
- Published: 2024-06-20
- Earliest claimed: 2022-12-14
- Assignees: Industrial Technology Research Institute
- Applicants: Industrial Technology Research Institute
- Inventors: Yu-Chi Lu; Po Hsuan Hsiao; Ming-Yen Chen; Chang Hong Lin; Hsin-Yeh Yang; Cheng-Hsuan Cheng
- Web page: https://patentgrep.com/patent/US20240202931A1

## CPC

- G06T 7/149: involving deformable models, e.g. active contour models
- G06T 7/62: of area, perimeter, diameter or volume
- G06V 10/44: Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06T 2207/30196: Human being; Person
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A measuring method and system for body-shaped data are provided. The measuring method comprises: performing a feature extraction on an apparel image to obtain a plurality of apparel feature points by a feature positioning module; performing a contour extraction on the apparel image to obtain an apparel boundary by an image segmentation module; calculating at least one shift value of the apparel feature points relative to the apparel boundary based on the apparel feature points and the apparel boundary by a processing module; correcting at least one of the apparel feature points according to the at least one shift value by the processing module, and projecting the corrected at least one of the apparel feature points to a three-dimensional model and obtaining at least one body-shaped data according to the apparel feature points on the three-dimensional model by the processing module.

## Claims

### Claim 1

A measuring method for body-shaped data, comprising:
performing, by a feature positioning module, a feature extraction on an apparel image to obtain a plurality of apparel feature points;
performing, by an image segmentation module, a contour extraction on the apparel image to obtain an apparel boundary;
calculating, by a processing module, at least one shift value of the apparel feature points relative to the apparel boundary based on the apparel feature points and the apparel boundary;
correcting, by the processing module, at least one of the apparel feature points according to the at least one shift value; and
projecting, by the processing module, the corrected at least one of the apparel feature points to a three-dimensional model, and obtaining at least one body-shaped data according to the apparel feature points on the three-dimensional model.

### Claim 2 (depends on 1)

The measuring method for body-shaped data according to claim 1, wherein performing the feature extraction on the apparel image to obtain the apparel feature points further comprises:
performing, by the feature positioning module, the feature extraction on a first type of apparel image and a second type of apparel image to obtain the apparel feature points corresponding to the first type of apparel image and the apparel feature points corresponding to the second type of apparel image; and
referring the first type of apparel image to the second type of apparel image, and rearranging the reference number of the apparel feature points of the first type of apparel image.

### Claim 3 (depends on 1)

The measuring method for body-shaped data according to claim 1, wherein calculating the at least one shift value of the apparel feature points relative to the apparel boundary based on the apparel feature points and the apparel boundary further comprises:
calculating, by the processing module, a shortest distance between an original position of each of the apparel feature points and the apparel boundary and taking the shortest distance as the at least one shift value.

### Claim 4 (depends on 3)

The measuring method for body-shaped data according to claim 3, wherein correcting at least one of the apparel feature points according to the at least one shift value comprising:
correcting, by the processing module, the apparel feature points respectively to be displaced from an original position of the apparel feature points by the shortest distance to a corrected position on the apparel boundary.

### Claim 5 (depends on 3)

The measuring method for body-shaped data according to claim 3, wherein the apparel feature point(s) whose shift value is zero is/are not be corrected, and projecting the corrected at least one of the apparel feature points to the three-dimensional model and obtaining at least one body-shaped data according to the apparel feature points on the three-dimensional model comprise:
projecting the corrected at least one of the apparel feature points and the uncorrected apparel feature point(s) to the three-dimensional model.

### Claim 6 (depends on 5)

The measuring method for body-shaped data according to claim 5, wherein projecting the corrected at least one of the apparel feature points to the three-dimensional model and obtaining at least one body-shaped data according to the apparel feature points on the three-dimensional model further comprise:
calculating, by the processing module, a distance between two of the apparel feature points in the three-dimensional model to obtain the at least one body-shaped data.

### Claim 7 (depends on 6)

The measuring method for body-shaped data according to claim 6, wherein the at least one body-shaped data includes a shoulder width, ae sleeve length, a garment length, a trouser length, a hip width or a skirt length.

### Claim 8 (depends on 1)

The measuring method for body-shaped data according to claim 1, wherein projecting the corrected at least one of the apparel feature points to the three-dimensional model and obtaining at least one body-shaped data according to the apparel feature points on the three-dimensional model further comprise:
calculating, by the processing module, a perimeter length of an intersection plane of the three-dimensional model where two of the apparel feature points are located to obtain the at least one body-shaped data.

### Claim 9 (depends on 1)

The measuring method for body-shaped data according to claim 1, wherein the apparel image on which the image segmentation module performs the contour extraction is a two-dimensional image.

### Claim 10 (depends on 1)

The measuring method for body-shaped data according to claim 1, wherein the apparel image is an image showing that a garment worn on a human body.

### Claim 11

A measuring system for body-shaped data, comprising:
a feature positioning module, configured to perform an feature extraction on an apparel image to obtain a plurality of apparel feature points;
an image segmentation module, configured to perform a contour extraction on the apparel image to obtain an apparel boundary; and
a processing module, configured to:
calculate at least one shift value of the apparel feature points relative to the apparel boundary based on the apparel feature points and the apparel boundary,
correct at least one of the apparel feature points according to the at least one shift value, and
project the corrected at least one of the apparel feature points to a three-dimensional model and obtain at least one body-shaped data according to the apparel feature points on the three-dimensional model.

### Claim 12 (depends on 11)

The measuring system for body-shaped data according to claim 11, wherein the feature positioning module is further configured to:
perform the feature extraction on a first type of apparel image and a second type of apparel image to obtain the apparel feature points corresponding to the first type of apparel image and the apparel feature points corresponding to the second type of apparel image; and
refer the first type of apparel image to the second type of apparel image and rearrange the reference number of the apparel feature points of the first type of apparel image.

### Claim 13 (depends on 11)

The measuring system for body-shaped data according to claim 11, wherein the processing module is configured to calculate a shortest distance between an original position of each of the apparel feature points and the apparel boundary and taking the shortest distance as the at least one shift value.

### Claim 14 (depends on 13)

The measuring system for body-shaped data according to claim 13, wherein the processing module is configured to correct the apparel feature points respectively to be displaced from an original position of the apparel feature points by the shortest distance to a corrected position on the apparel boundary.

### Claim 15 (depends on 13)

The measuring system for body-shaped data according to claim 13, wherein the processing module does not correct the apparel feature point(s) whose shift value is zero, the processing module is configured to project the corrected at least one of the apparel feature points and the uncorrected apparel feature point(s) to the three-dimensional model.

### Claim 16 (depends on 11)

The measuring system for body-shaped data according to claim 11, wherein the processing module is configured to calculate a distance between two of the apparel feature points in the three-dimensional model to obtain the at least one body-shaped data.

### Claim 17 (depends on 16)

The measuring system for body-shaped data according to claim 16, wherein the at least one body-shaped data includes a shoulder width, a sleeve length, a garment length, a trouser length, a hip width or a skirt length.

### Claim 18 (depends on 11)

The measuring system for body-shaped data according to claim 11, wherein the processing module is configured to calculate a perimeter length of an intersection plane of the three-dimensional model where two of the apparel feature points are located to obtain the at least one body-shaped data.

### Claim 19 (depends on 11)

The measuring system for body-shaped data according to claim 11, wherein the apparel image on which the image segmentation module performs the contour extraction is a two-dimensional image.

### Claim 20 (depends on 11)

The measuring system for body-shaped data according to claim 11, wherein the apparel image is an image showing that a garment worn on a human body.
