Automatic snap for digital sketch inking

Patent metadata

Field	Value
Publication number	US-10210636-B2
Application number	US-201615253540-A
Country	US
Kind code	B2
Filing date	Aug 31, 2016
Priority date	Aug 31, 2016
Publication date	Feb 19, 2019
Grant date	Feb 19, 2019

Abstract

A method, system, and computer program product provide the ability to ink a drawing. A coarse original sketch is loaded. User input drawing an initial stroke over the coarse original sketch is accepted. A stroke smoothness of the initial stroke is measured. A similarity between the initial stroke and the coarse original sketch is measured to determine a snap affect. The stroke smoothness and the snap affect are combined. The initial stroke is automatically/dynamically, without additional user input, modified based on the combination.

First claim

What is claimed is:

1. A computer-implemented method for inking a drawing comprising:
(a) loading, into a computer, a coarse original sketch;
(b) accepting, into the computer, user input drawing an initial stroke over the coarse original sketch;
(c) measuring, in the computer, a stroke smoothness of the initial stroke;
(d) measuring, in the computer, a similarity between the initial stroke and the coarse original sketch to determine a snap effect, wherein the measuring the similarity comprises:
(1) acquiring sampled points in the initial stroke;
(2) determining a local compact region of the coarse original sketch for each of the sampled points;
(3) computing distance decaying weights for each of the sampled points based on the local compact region;
(4) computing a drag force for each of the sampled points based on the distance decaying weights; and
(5) computing a total snap energy term that determines how pixels in the coarse original sketch affect the initial stroke, wherein:
the total snap energy term (E snap ) comprises:
E snap =Σ i f i =Σ i Σ q∈N(p i ) w q ·(255 −I q )·( q−p i ) 2 ;

wherein q is a pixel in the local compact region around sampled point p i represented as N(p i ), w q is the distance decaying weight for sampled point p i , I q is an intensity of pixel q in the coarse original sketch, and f i represents the drag force;

(e) combining, in the computer, the stroke smoothness and the snap effect; and
(f) automatically, without additional user input, modifying and displaying, via the computer, the initial stroke based on the combining.

2. The computer-implemented method of claim 1 , wherein the loading comprises:
scanning a paper drawing to generate a raster image in the computer.

3. The computer-implemented method of claim 1 , wherein the measuring the stroke smoothness comprises:
acquiring sampled points in the initial stroke;
storing the sampled points;
computing one or more turning vectors for the sampled points;
computing a smooth energy term that minimizes total changes of the turning vectors.

4. The computer-implemented method of claim 3 , wherein:
the sampled points comprise p 1 , . . . p n ;
the turning vector comprises a change of edge vectors at position p i :
κ i =( p i −p i−1 )−( p i+1 −p i )=2 p i −p i−1 −p i+1 ; and

the smooth energy term comprises:
E smooth =Σ i (κ i+1 −κ i ) 2 .

5. The computer-implemented method of claim 1 , wherein:
the local compact region comprises a rotated two dimensional (2D) Gaussian.

6. The computer-implemented method of claim 1 , wherein:
the distance decaying weights are computed based on:

w
⁡

(

u
,
v

)

=

Ae

-

[

(

u
-

u
0

)

2

2
⁢

σ
u
2

+

(

v
-

v
0

)

2

2
⁢

σ
v
2

]

wherein A is an aptitude, (u,v) is a local coordinate at position q=q(x,y) on the initial stroke where a frame axis is aligned with a tangent vector {right arrow over (u)} and binormal {right arrow over (v)} vector, σ u , σ v control a shape of the local compact region in tangent and binormal directions respectively.

7. The computer-implemented method of claim 1 , wherein:
the drag force for each sampled point p i is computed by:
f i Σ q∈N(p i ) w q ·(255 −I q )·( q−p i ) 2

wherein q is a pixel in the local compact region around p i represented as N(p i ), w q is the distance decaying weight for sampled point p i , and I q is an intensity of pixel q in the coarse original sketch.

8. The computer-implemented method of claim 1 , wherein the combining computes a final Energy (E):
E=E smooth +λE snap
wherein E smooth is the stroke smoothness, and A is a parameter to tune a strength of the snap effect (E snap );
the final energy is solved using a linear solver, by zeroing each unknown variables' first order partial derivatives:

∂
E

∂

p
i

=
0

,

i
=

1
⁢

⁢
…
⁢

⁢

n
.

9. A system for inking a drawing in a computer system comprising:
(a) a computer having a memory;
(b) an application executing on the computer, wherein the application:
(1) loads a coarse original sketch;
(2) accepts user input drawing an initial stroke over the coarse original sketch;
(3) measures a stroke smoothness of the initial stroke;
(4) measures a similarity between the initial stroke and the coarse original sketch to determine a snap effect, wherein the measuring the similarity comprises:
(i) acquiring sampled points in the initial stroke;
(ii) determining a local compact region of the coarse original sketch for each of the sampled points;
(iii) computing distance decaying weights for each of the sampled points based on the local compact region;
(iv) computing a drag force for each of the sampled points based on the distance decaying weights;

Assignees

• Autodesk Inc

Inventors

• Liu Ning
• Wang Lei

Classifications

• G06T11/23 Primary

  using straight lines or curves · CPC title

• G06T3/0081

  Physics · mapped topic

• G06T11/203 Primary

  Physics · mapped topic

• G06T3/606

  by memory addressing or mapping · CPC title

• G06T3/153

  using elastic snapping · CPC title