Granted Patent B2

US 12,462,515 · App. 18/541,479 · Granted Nov 4, 2025

Inventors: Laurent Juppe (Montreal, CA); Sherif Esmat Omar Abuelwafa (Montreal, CA)

Assignee: APPLICATIONS MOBILES OVERVIEW INC.

G06T19/20G06T17/00G06T2210/56G06T2219/2004G06T2219/2016

Abstract

Claims (60)

1 . A method of generating a three-dimensional (3D) representation of a foot of a user, the method comprising:

capturing data of the foot using one or more cameras;

generating, based on the data of the foot, a 3D point cloud corresponding to the foot;

retrieving a 3D model template of a generic human foot;

generating the 3D representation of the foot by:

aligning the 3D point cloud to the 3D model template using an iterative closest point (ICP) algorithm, and

applying local deformations to a surface of the 3D model template by:

determining a normal axis extending from a point of the 3D model template;

determining a plurality of neighboring points located within a radius of the point of the 3D model template;

determining a centroid of the plurality of neighboring points;

displacing the point along the normal axis; and

displacing the neighboring points based on a product of the normal axis and a vector linking a vertex to be displaced and the centroid; and

determining, based on the 3D representation of the foot, measurement information of the foot.

2 . The method of claim 1 , wherein the measurement information comprises a length, height, depth, or width of the foot.

3 . The method of claim 1 , wherein the aligning the 3D point cloud to the 3D model template further comprises adjusting a scale of the 3D model template using the ICP algorithm.

4 . The method of claim 1 , wherein aligning the 3D point cloud to the 3D model template is performed iteratively and comprises:

modifying a scale of the 3D model template, applying or refining a rotation to the 3D model template, or applying or refining a translation to the 3D model template; and

calculating a distance between the 3D model template and the 3D point cloud.

5 . The method of claim 1 , wherein capturing the data of the foot comprises capturing images of the foot or a video of the foot.

6 . The method of claim 1 , wherein the one or more cameras are embedded in a mobile phone device.

7 . The method of claim 1 , further comprising selecting the 3D model template from a database of 3D models.

8 . The method of claim 1 , wherein the 3D model template comprises a 3D mesh model.

9 . The method of claim 8 , further comprising generating a 3D point cloud model template based on the 3D mesh model, and wherein aligning the 3D point cloud to the 3D model template comprises aligning the 3D point cloud to the 3D point cloud model template.

10 . The method of claim 1 , further comprising applying a denoising routine to the data of the foot.

11 . The method of claim 1 , wherein aligning the 3D point cloud to the 3D model template comprises:

computing 2D pictures from the 3D model template in a combination of different points of view; and

matching the 2D pictures to the different points of view.

12 . A system comprising at least one processor and memory storing executable instructions which, when executed by the at least one processor, cause the system to:

receive data of a foot captured using one or more cameras;

generate, based on the data of the foot, a 3D point cloud corresponding to the foot;

retrieve a 3D model template of a generic human foot;

generate a 3D representation of the foot by:

aligning the 3D point cloud to the 3D model template using an iterative closest point (ICP) algorithm to generate a 3D representation of the foot, and

applying local deformations to a surface of the 3D model template by:

determining a normal axis extending from a point of the 3D model template;

determining a plurality of neighboring points located within a radius of the point of the 3D model template;

determining a centroid of the plurality of neighboring points;

displacing the point along the normal axis; and

displacing the neighboring points based on a product of the normal axis and a vector linking a vertex to be displaced and the centroid; and

determine, based on the 3D representation of the foot, measurement information of the foot.

13 . The system of claim 12 , wherein the measurement information comprises a length, height, depth, or width of the foot.

14 . The system of claim 12 , wherein the instructions that cause the system to align the 3D point cloud to the 3D model template comprise instructions that cause the system to adjust a scale of the 3D model template using the ICP algorithm.

15 . The system of claim 12 , wherein the instructions that cause the system to align the 3D point cloud to the 3D model template comprise instructions that cause the system to:

modify a scale of the 3D model template, applying or refining a rotation to the 3D model template, or applying or refining a translation to the 3D model template; and

calculate a distance between the 3D model template and the 3D point cloud.

16 . The system of claim 12 , wherein the 3D model template comprises a 3D mesh model, and the instructions further cause the system to generate a 3D point cloud model template based on the 3D mesh model, and wherein the instructions that cause the system to align the 3D point cloud to the 3D model template comprise instructions that cause the system to align the 3D point cloud to the 3D point cloud model template.

17 . A non-transitory computer-readable medium comprising executable instructions which, when executed by at least one processor, cause the at least one processor to:

receive data of a foot captured using one or more cameras;

generate, based on the data of the foot, a 3D point cloud corresponding to the foot;

retrieve a 3D model template of a generic human foot;

generate a 3D representation of the foot by:

aligning the 3D point cloud to the 3D model template using an iterative closest point (ICP) algorithm to generate a 3D representation of the foot, and

applying local deformations to a surface of the 3D model template by:

determining a normal axis extending from a point of the 3D model template;

determining a plurality of neighboring points located within a radius of the point of the 3D model template;

determining a centroid of the plurality of neighboring points;

displacing the point along the normal axis; and

displacing the neighboring points based on a product of the normal axis and a vector linking a vertex to be displaced and the centroid; and

determine, based on the 3D representation of the foot, measurement information of the foot.

18 . The non-transitory computer-readable medium of claim 17 , wherein the measurement information comprises a length, height, depth, or width of the foot.

Assignments (2)

Continuity (4)

References Cited (50)