IP Library › Granted Patent US 9,251,620
Granted Patent B2
US 9,251,620 · App. 13/908,969 · Granted Feb 2, 2016

Fast pattern interpolation with parallel coherent random walk searching

Inventors: Christian Eisenacher (Burbank, CA); Svetlana Akim (Ebelsbach, DE)
Assignee: Disney Enterprises, Inc.
G06T15/04G06T11/001
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,251,620
App. No.
13/908,969
Filed
Jun 3, 2013
Granted
Feb 2, 2016
Kind
B2
Art Unit
2619
USPC
345/419
Abstract

There is provided a method and system for fast pattern interpolation with parallel coherent random walk searching. The system includes a system processor and a system memory, and an image processing application stored in the system memory. The image processing application, under control of the system processor, is configured to receive first and second pattern exemplars and a blend map for merging the first and second pattern exemplars, and to utilize a correspondence search process to determine first and second target images for use in producing a composite image. The first and second target images correspond respectively to the first and second pattern exemplars. In addition, a constraint imposed on the correspondence search process by the image processing application is based on the blend map for merging the first and second pattern exemplars.

Claims (44)

1. A method for image processing, the method comprising:

receiving first and second pattern exemplars and a blend map for merging the first and second pattern exemplars; and

utilizing a correspondence search process, including a parallel coherent random walk search, to determine first and second target images for use in producing a composite image, the first and second target images corresponding respectively to the first and second pattern exemplars;

wherein a constraint imposed on the correspondence search process is based on the blend map for merging the first and second pattern exemplars.

2. The method of claim 1 , further comprising merging the first and second target images to produce the composite image.

3. The method of claim 1 , wherein the composite image comprises a texture composition.

4. The method of claim 1 , further comprising merging the first and second target images to produce the composite image as an input exemplar for use in a texture synthesis process.

5. The method of claim 4 , wherein the texture synthesis process is implemented using a texture by numbers (TBN) process.

6. The method of claim 4 , further comprising synthesizing a texture composition including a first pattern corresponding to the first pattern exemplar, a second pattern corresponding to the second pattern exemplar, and transitions between the first and second patterns generated in the composite image.

7. The method of claim 1 , further comprising synthesizing and merging the first and second target images to produce the composite image on a three-dimensional (3D) surface.

8. The method of claim 1 , further comprising mapping the first pattern exemplar and the second pattern exemplar to a shared appearance space prior to utilizing the correspondence search process to determine the first and second target images.

9. A system for performing image processing, the system comprising:

a system processor and a system memory;

an image processing application stored in the system memory, the image processing application, under control of the system processor, configured to:

receive first and second pattern exemplars and a blend map for merging the first and second pattern exemplars; and

utilize a correspondence search process, a parallel coherent random walk search, to determine first and second target images for use in producing a composite image, the first and second target images corresponding respectively to the first and second pattern exemplars;

wherein a constraint imposed on the correspondence search process by the image processing application is based on the blend map for merging the first and second pattern exemplars.

10. The system of claim 9 , wherein the image processing application is configured to merge the first and second target images to produce the composite image.

11. The system of claim 9 , wherein the composite image comprises a texture composition.

12. The system of claim 9 , wherein the image, processing application is further configured to merge the first and second target images to produce the composite image as an input exemplar for use in a texture synthesis process.

13. The system of claim 12 , wherein the image processing application is further configured to synthesize a texture composition including a first pattern corresponding to the first pattern exemplar, a second pattern corresponding to the second pattern exemplar, and transitions between the first and second patterns generated in the composite image.

14. The system of claim 9 , wherein the image processing application is further configured to synthesize and merge the first and second target images to produce the composite image on a three-dimensional (3D) surface.

15. The system of claim 9 , wherein the image processing application is further configured to map the first pattern exemplar and the second pattern exemplar to a shared appearance space prior to utilizing the correspondence search process to determine the first and second target images.

16. A non-transitory computer readable memory having an image processing application stored therein for execution by a processor for generating a processed image by a method comprising:

receiving first and second pattern exemplars and a blend map for merging the first and second pattern exemplars; and

utilizing a correspondence search process, including a parallel coherent random walk search, to determine first and second target images for use in producing a composite image, the first and second target images corresponding respectively to the first and second pattern exemplars;

wherein a constraint imposed on the correspondence search process is based on the blend map for merging the first and second pattern exemplars.

17. A method for image processing, the method comprising:

receiving first and second pattern exemplars and a blend map for merging the first and second pattern exemplars; and

utilizing a correspondence search process to determine first and second target images for use in producing a composite image, the first and second target images corresponding respectively to the first and second pattern exemplars;

wherein the correspondence search process includes a first series of parallel coherent random walk searches for determining the first target image and a second series of parallel coherent random walk searches for determining the second target image, the first series and the second series of parallel coherent random walk searches being performed substantially concurrently, and wherein a constraint imposed on the correspondence search process is based on the blend map for merging the first and second pattern exemplars.

18. The method of claim 17 further comprising:

mapping the first pattern exemplar and the second pattern exemplar to a shared appearance space prior to utilizing the correspondence search process to determine the first and second target images; and

merging the first and second target images to produce the composite image.

19. A system for performing image processing, the system comprising:

a system processor and a system memory;

an image processing application stored in the system memory, the image processing application, under control of the system processor, configured to:

receive first and second pattern exemplars and a blend map for merging the first and second pattern exemplars; and

utilize a correspondence search process to determine first and second target images for use in producing a composite image, the first and second target images corresponding respectively to the first and second pattern exemplars;

wherein the correspondence search process utilized by the image processing application includes a first series of parallel coherent random walk searches for determining the first target image and a second series of parallel coherent random walk searches for determining the second target image, the first series and the second series of parallel coherent random walk searches being performed substantially concurrently, and wherein a constraint imposed on the correspondence search process by the image processing application is based on the blend map for merging the first and second pattern exemplars.

20. A non-transitory computer readable memory having an image processing application stored therein for execution by a processor for generating a processed image by a method comprising:

receiving first and second pattern exemplars and a blend map for merging the first and second pattern exemplars; and

utilizing a correspondence search process to determine first and second target images for use in producing a composite image, the first and second target images corresponding respectively to the first and second pattern exemplars;

wherein the correspondence search process includes a first series of parallel coherent random walk searches for determining the first target image and a second series of parallel coherent random walk searches for determining the second target image, the first series and the second series of parallel coherent random walk searches being performed substantially concurrently, and wherein a constraint imposed on the correspondence search process is based on the blend map for merging the first and second pattern exemplars.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 7, 2013
From: EISENACHER, CHRISTIAN; AKIM, SVETLANA
To: DISNEY ENTERPRISES, INC
Reel/Frame 030569/0452 →
Continuity (2)
Continuation In Part 13281838 · Oct 26, 2011
Related Publication 20140015828A1 · Jan 16, 2014