Techniques for improving stereo block matching with the pyramid method

Patent metadata

Field	Value
Publication number	US-9396408-B2
Application number	US-201113976907-A
Country	US
Kind code	B2
Filing date	Dec 15, 2011
Priority date	Dec 15, 2011
Publication date	Jul 19, 2016
Grant date	Jul 19, 2016

Abstract

Techniques to determine a search range for a stereo based matching pyramid. A first disparity estimation value for a first level in a stereo based matching pyramid based on an image may be received. A search range for a second level may be determined using the first disparity estimation value. The search range based on a pyramid level of a second level may be increased. The search range may be increased based on a pyramid level of the second level. A second disparity estimation value may be selected from the search area for the second level. A depth map for the second level may be determined based on the second disparity estimation value. Other embodiments are described and claimed.

First claim

The invention claimed is:

1. An article of manufacture comprising a storage medium containing instructions that when executed cause a system to:
receive a first disparity estimation value for a first level in a stereo based matching pyramid based on an image;
determine a search range for a second level based on the first disparity estimation value;
increase the search range based on a pyramid level of the second level; and
select a second disparity estimation value from the search range for the second level based on a sum of absolute differences on a gradient times an absolute difference of a gray.

2. The article of manufacture of claim 1 , comprising instructions that when executed cause the system to determine a depth map for the second level based on the second disparity estimation value.

3. The article of manufacture of claim 1 , comprising instructions that when executed cause the system to:
calculate a Laplacian value; and
determine the search range within the second level based on the Laplacian value.

4. The article of manufacture of claim 1 , comprising instructions that when executed cause the system to select the second disparity estimation value based on a correlation score.

5. The article of manufacture of claim 1 , comprising instructions that when executed cause the system to determine that a second level image on the second level includes a thin object that was not included on a first level image on the first level.

6. The article of manufacture of claim 1 , comprising instructions that when executed cause the system to select the second disparity estimation value based on a local binary pattern.

7. An apparatus, comprising:
a processing component;
a pyramid level search range component operatively coupled to the processing component to:
determine a search range for a second level based on a first disparity estimation value from a first level;
increase the search range based on a pyramid level of the second level; and
select a second disparity estimation value from the search range for the second level; and

a correlation component to determine a correlation score based on a sum of absolute differences on a gradient times an absolute difference of a gray.

8. The apparatus of claim 7 , comprising:
a digital display operatively coupled to the processing component.

9. The apparatus of claim 7 , comprising:
a pixel search range component to:
calculate a Laplacian value; and
determine the search range within the second level based on the Laplacian value.

10. The apparatus of claim 7 , comprising:
a pixel search range component to:
determine that a target object is in the search range for the second level;
determine that the target object is not presented in the first level; and
increase the search range within the second level.

11. The apparatus of claim 7 , comprising:
a pixel search range component to increase the search range within the second level when the search range comprises edges.

12. The apparatus of claim 7 , comprising:
a correlation component to determine a correlation score based on a local binary pattern.

13. A method, comprising:
receiving, at a processing component, a first image from a first camera and a second image from a second camera and from a camera;
determining, by the processing component, a stereo based matching pyramid from at least one of the first and second images;
receiving a first disparity estimation value for a first level in the stereo based matching pyramid;
determining, by the processing component, a search range for a second level based on the first disparity estimation value;
increasing the search range based on a pyramid level of a second level;
selecting, by the processing component, a second disparity estimation value from the search range for the second level;
determining, by the processing component, a depth map for the second level based on the second disparity estimation value; and
determining, by the processing component, a correlation score based on a sum of absolute differences on a gradient times an absolute difference of a gray.

14. The method of claim 13 , comprising:
calculating a Laplacian value; and
determining the search range within the second level based on the Laplacian value.

15. The method of claim 13 , comprising:
selecting a second disparity estimation value from the search range for the second level based on a correlation score.

16. The method of claim 13 , comprising:
increasing the search range within the second level when the search range comprises edges.

17. The method of claim 13 , comprising:
determining a correlation score based on a local binary pattern.

18. A system, comprising:
a processing component;
a memory to store a depth extraction application;
an operating system to load the stereo block matching pyramid application on the processing component, the stereo block matching pyramid application operative on the processing component to:
determine a search range for a second level based on the first disparity estimation value;
increase the search range based on a pyramid level of a second level;
select a second disparity estimation value from the search range for the second level, and
determine a depth map for the second level based on the second disparity estimation value;
an interface to communicate information between the processing component and the operating system; and

determine a correlation score based on a sum of absolute differences on a gradient times an absolute difference of a gray.

19. The system of claim 18 , comprising a pixel search range component in the stereo block matching application operating on the processing component to:
determine that a target object is in the search range for the second level;
determine that the target object is not presented in the first level; and
increase the search range within the second level.

20. The system of claim 18 , comprising a pixel search range component in the stereo block matching application operating on the processing component to:
increase the search range within the second level when the search range comprises edges.

21. The system of claim 18 , comprising a correlation component in the stereo block matching application operating on the processing component to:
determine a correlation score based on a local binary pattern.

Assignees

• Reif Dror
• Aviv Ziv
• Stanhill David
• Intel Corp

Inventors

• Reif Dror
• Aviv Ziv
• Stanhill David

Classifications

• G06T7/0051

  Physics · mapped topic

• G06T2207/20016

  Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform · CPC title

• G06T7/0075

  Physics · mapped topic

• G06T2207/10012

  Stereo images · CPC title

• H04N2013/0074

  Stereoscopic image analysis · CPC title