Extraction of body dimensions from planar garment photographs of fitting garments

• US 10,410,414 B2
• Filed: 08/03/2018
• Issued: 09/10/2019
• Est. Priority Date: 11/14/2013
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Su, Jonathan, Naware, Mihir, Chhugani, Jatin
• Primary Examiner(s): Demeter, Hilina K
• Application Number:
  US16/054,139
• Publication Number:
  US 20180350140A1
• Time in Patent Office: 403 Days
• Field of Search: None
• CPC Class Codes: A41H 1/00 Measuring aids or methods
• CPC Class Codes: A41H 3/007 Methods of drafting or marking-out patterns using computers computer-aided design G06F30/00
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: G06F 2111/02 CAD in a network environment, e.g. collaborative CAD or distributed simulation
• CPC Class Codes: G06F 2113/12 Cloth
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06Q 30/0643 Graphical representation of items or shoppers
• CPC Class Codes: G06T 15/005 General purpose rendering architectures
• CPC Class Codes: G06T 17/00 Three dimensional [3D] modelling, e.g. data description of 3D objects
• CPC Class Codes: G06T 17/10 Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
• CPC Class Codes: G06T 17/20 Finite element generation, e.g. wire-frame surface description, tesselation
• CPC Class Codes: G06T 19/00 Manipulating 3D models or images for computer graphics
• CPC Class Codes: G06T 19/20 Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
• CPC Class Codes: G06T 2210/16 Cloth
• CPC Class Codes: G06T 2215/16 Using real world measurements to influence rendering

Abstract

Techniques for extraction of body parameters, dimensions and shape of a customer are presented herein. A model descriptive of a garment, a corresponding calibration factor and reference garment shapes can be assessed. A garment shape corresponding to the three-dimensional model can be selected from the reference garment shapes based on a comparison of the three-dimensional model with the reference garment shapes. A reference feature from the plurality of reference features may be associated with the model feature. A measurement of the reference feature may be calculated based on the association and the calibration factor. The computed measurement can be stored in a body profile associated with a user. An avatar can be generated for the user based on the body profile and be used to show or indicate fit of a garment, as well as make fit and size recommendations.

Claims

1. A method comprising:

comparing, by a computer system, a three-dimensional model descriptive of a garment to reference garment shapes for different types of garments;

selecting, by the computer system, a garment shape from the reference garment shapes corresponding to the three-dimensional model descriptive of the garment;

comparing, by the computer system, a model feature of the three-dimensional model descriptive of the garment to reference features of the selected garment shape to determine a reference feature of the selected garment shape that is similar to the model feature of the three-dimensional model descriptive of the garment;

computing, by the computer system, a measurement of the reference feature based on the model feature using a calibration factor indicating the size of the model feature;

storing, by the computer system, the computed measurement of the reference feature in a body profile associated with a user that comprises one or more computed measurements of reference features; and

generating an avatar for the user based on the body profile including the computed measurement of the reference feature.

2. The method of claim 1, wherein generating the avatar further comprises:

selecting a first avatar from a database of reference avatars based on the body profile, the first avatar having a feature measurement less than the computed measurement of the reference feature;

selecting a second avatar from the database of reference avatars, the second avatar having a feature measurement more than the computed measurement of the reference feature; and

generating the avatar for the user by interpolating the first avatar and the second avatar.

3. The method of claim 2, wherein selecting the first avatar comprises:

computing a difference vector for each of a subset of avatars in the database of reference avatars, each difference vector being based on the computed measurement of the reference feature and the feature measurement of each avatar in the subset of avatars; and

selecting the first avatar from the subset of avatars based on the difference vector of the first avatar being less than the difference vector of remaining avatars in the subset of avatars.

4. The method of claim 2, wherein the interpolating the first avatar and the second avatar includes:

linearly interpolating between the feature measurement of the first avatar and the feature measurement of the second avatar.

5. The method of claim 1, wherein the body profile includes a rise measurement, an inseam measurement, a waist measurement, a chest measurement, a thigh measurement, or a calf measurement.

6. The method of claim 1, wherein the body profile includes a gender indicator that corresponds to the user, and wherein measurements stored in the body profile differ based on the gender indicator.

7. The method of claim 1, wherein the three-dimensional model includes material properties, and the method further comprises:

presenting, on a display of a device, the three-dimensional model with a fit map based on the material properties and the body profile, the fit map having force data associated with a fit of the garment.

8. The method of claim 1, further comprising:

presenting, on a display of a device, the three-dimensional model draped on the avatar for the user.

9. The method of claim 1, wherein the three-dimensional model is based on multiple photographs of the garment, and the calibration factor is generated based on an object of a pre-determined size in a photograph from the multiple photographs.

10. The method of claim 9, wherein the size for the model feature is calculated by measuring the model feature in a photograph from the multiple photographs, and multiplying the measured model feature by the calibration factor.

11. The method of claim 1, wherein the calibration factor is different for different zones of the garment.

12. The method of claim 1, wherein the three-dimensional model is based on multiple photographs of the garment, and the calibration factor is based on a photograph from the one or more photographs having a tiled background with a pre-determined size of tile.

13. The method of claim 1, further comprising:

determining a size from a set of sizes for another garment based on the body profile associated with the user.

14. The method of claim 1, wherein the body profile includes a fit preference based on a purchase history of the user.

15. A system comprising:

a memory that stores instructions; and

one or more processors configured by the instructions to perform operations comprising;

comparing a three-dimensional model descriptive of a garment to reference garment shapes for different types of garments;

selecting a garment shape from the garment shapes corresponding to the three-dimensional model descriptive of the garment;

comparing a model feature of the three-dimensional model descriptive of the garment to reference features of the selected garment shape to determine a reference feature of the selected garment shape that is similar to the model feature of the three-dimensional model descriptive of the garment;

computing a measurement of the reference feature based on the model feature using a calibration factor indicating the size of the model feature;

storing the computed measurement of the reference feature in a body profile associated with a user that comprises one or more computed measurements of reference features; and

generating an avatar for the user based on the body profile including the computed measurement of the reference feature.

16. The system of claim 15, wherein generating the avatar further comprises:

selecting a first avatar from a database of reference avatars based on the body profile, the first avatar having a feature measurement less than the computed measurement of the reference feature;

selecting a second avatar from the database of reference avatars, the second avatar having a feature measurement more than the computed measurement of the reference feature; and

generating the avatar for the user by interpolating the first avatar and the second avatar.

17. The system of claim 16, wherein selecting the first avatar comprises:

computing a difference vector for each of a subset of avatars in the database of reference avatars, each difference vector being based on the computed measurement of the reference feature and the feature measurement of each avatar in the subset of avatars; and

selecting the first avatar from the subset of avatars based on the difference vector of the first avatar being less than the difference vector of remaining avatars in the subset of avatars.

18. The system of claim 16, wherein the interpolating the first avatar and the second avatar includes:

linearly interpolating between the feature measurement of the first avatar and the feature measurement of the second avatar.

19. The system of claim 15, wherein the body profile includes at least one of a rise measurement, an inseam measurement, a waist measurement, a chest measurement, a thigh measurement, and a calf measurement.

20. A non-transitory machine-readable storage medium comprising instructions that when executed by one or more processors of a machine, cause the machine to perform operations comprising:

comparing a three-dimensional model descriptive of a garment to reference garment shapes for different types of garments;

selecting a garment shape from the garment shapes corresponding to the three-dimensional model descriptive of the garment;

comparing a model feature of the three-dimensional model descriptive of the garment to reference features of the selected garment shape to determine a reference feature of the selected garment shape that is similar to the model feature of the three-dimensional model descriptive of the garment;

computing a measurement of the reference feature based on the model feature using a calibration factor indicating the size of the model feature;

storing the computed measurement of the reference feature in a body profile associated with a user that comprises one or more computed measurements of reference features; and

generating an avatar for the user based on the body profile including the computed measurement of the reference feature.