Auto proxy mesh generation pipeline for planar-shape cloth

Patent metadata

Field	Value
Publication number	US-12530858-B2
Application number	US-202318383790-A
Country	US
Kind code	B2
Filing date	Oct 25, 2023
Priority date	Oct 25, 2023
Publication date	Jan 20, 2026
Grant date	Jan 20, 2026

Abstract

A proxy mesh that includes a plurality of proxy vertices is generated by projecting a visual mesh that includes a plurality of visual vertices onto a fitting plane. The visual mesh is of a character model. Each of the plurality of proxy vertices indicates a respective bone of the character model. A vertex weight associated with each of the plurality of proxy vertices of the proxy mesh is determined. Each of the vertex weights indicates how much the respective proxy vertex impacts a position of a visual vertex of the plurality of visual vertices that is adjacent to the respective proxy vertex. A position of each of the plurality of visual vertices is adjusted based on the determined vertex weights of a subset of the plurality of proxy vertices that are adjacent to the respective visual vertex.

First claim

What is claimed is:

1 . A method of mesh processing, the method comprising:
generating a proxy mesh that includes a plurality of proxy vertices, the proxy mesh being generated based on a visual mesh by projecting the visual mesh that includes a plurality of visual vertices onto a fitting plane, the visual mesh being of an object worn by a character model, each of the plurality of proxy vertices indicating a respective bone of the character model and neighboring one or more of the plurality of visual vertices of the visual mesh; determining a vertex weight associated with each of the plurality of proxy vertices of the proxy mesh, each of the vertex weights indicating how much the respective proxy vertex impacts a position of one of the one or more of the plurality of visual vertices that is adjacent to the respective proxy vertex; and adjusting a position of each of the plurality of visual vertices based on the determined vertex weights of a subset of the plurality of proxy vertices that are adjacent to the respective visual vertex.

2 . The method of claim 1 , wherein the generating the proxy mesh further comprises:
determining a first principal component of the plurality of visual vertices based on a principal component analysis (PCA), the first principal component including a most variation of the plurality of visual vertices; and determining the fitting plane based on the first principal component in which a normal vector of the fitting plane is determined based on a direction of the first principal component.

3 . The method of claim 2 , wherein the generating the proxy mesh further comprises:
obtaining samples from each of a plurality of faces of the visual mesh based on a uniform sampling, each of the plurality of faces of the visual mesh being defined by a respective subset of the plurality of visual vertices; projecting the obtained samples from the plurality of faces of the visual mesh onto the fitting plane; and determining a two-dimensional (2D) signed distance field on the fitting plane on which the obtained samples are projected.

4 . The method of claim 3 , wherein the generating the proxy mesh further comprises:
extracting a plurality of 2D isolines from the 2D signed distance field based on an iso-value through Marching Squares, an outer boundary of the 2D signed distance field being defined by the plurality of 2D isolines.

5 . The method of claim 4 , wherein the extracting the plurality of 2D isolines further comprises:
determining a plurality of 2D squares in the 2D signed distance field based on samples in the 2D signed distance field, each of the plurality of 2D squares including 4 respective samples; classifying states of the samples of each of the plurality of 2D squares based on the iso-value, the states of the samples of the respective one of the plurality of 2D squares indicating whether the samples are larger than the iso-value; determining a point between each pair of adjacent samples of the plurality of 2D squares that have different states; and connecting adjacent points of the determined points to form the plurality of 2D isolines.

6 . The method of claim 5 , wherein the generating the proxy mesh further comprises:
generating a plurality of simplified line segment loops based on the plurality of 2D isolines, the plurality of simplified line segment loops being generated based on a subset of the determined points, a distance between the plurality of simplified line segment loops and the plurality of 2D isolines being equal to or less than a threshold value.

7 . The method of claim 6 , wherein the generating the proxy mesh further comprises:
generating a Poisson disk sample set that includes a plurality of samples within a domain defined by the plurality of simplified line segment loops in which a distance between each pair of neighboring samples of the Poisson disk sample set is larger than a threshold distance; and removing a subset of the plurality of samples to generate the plurality of proxy vertices, the subset of the plurality of samples being removed based on a weighted sample elimination in which each of the plurality of samples is assigned with a weight and the subset of the plurality of samples are removed according to the respective weights.

8 . The method of claim 7 , wherein the removing further comprises:
assigning each of the plurality of samples with the respective weight; determining a first sample of the plurality of samples and neighboring samples of the first sample based on a k-dimensional tree structure of the plurality of samples; building a heap based on the first sample of the plurality of samples and the neighboring samples of the first sample, the heap indicating that the first sample and the neighboring samples of the first sample are organized in a binary tree-based structure based on the weights of the first sample and the neighboring samples of the first sample; and removing one of the neighboring samples of the first sample that corresponds to a largest weight in the heap.

9 . The method of claim 7 , wherein the weight of each sample of the plurality of samples is associated with distances between the respective sample and respective neighboring samples of the respective sample.

10 . The method of claim 1 , wherein the generating the proxy mesh further comprises:
generating the proxy mesh based on a plurality of Delaunay triangles that are formed based on the plurality of proxy vertices, a circumcircle of each of the plurality of Delaunay triangles being an empty circle.

11 . The method of claim 1 , wherein the adjusting the position of each of the plurality of visual vertices further comprises:
adjusting the position of a i-th visual vertex of the plurality of visual vertices based on four proxy vertices of the plurality of proxy vertices adjacent to the i-th visual vertex according to v′ i =(Σ j=1 4 w i,j T j ) v i , v i being an initial position of the i-th visual vertex, v′ i being the adjusted position of the i-th visual vertex, T j being a spatial transformation matrix associated with a j-th proxy vertex of the four proxy vertices, w i,j being the vertex weight that is associated with the j-th proxy vertex and indicates how much the j-th proxy vertex impacts the initial position of the i-th visual vertex.

12 . The method of claim 11 , wherein the determining the vertex weight further comprises:
determining the vertex weight w i,j associated with the j-th proxy vertex as:

w

i
,
j

=

⁢

{

1
,

if
⁢

⁢

d
j

/

d
max

<

Assignees

• Tencent America LLC

Inventors

• Wu Kui
• Pan Zherong
• Gao Xifeng
• Cai Guangyi
• Feng Qijia
• Wang Lanye

Classifications

• G06T7/13

  Edge detection · CPC title

• G06T2219/2004

  Aligning objects, relative positioning of parts · CPC title

• G06T2210/16

  Cloth · CPC title

• G06T17/20

  Finite element generation, e.g. wire-frame surface description, {tesselation} · CPC title

• G06T19/20 Primary

  Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts · CPC title