Object Selection in Stereo Image Pairs

• US 20130127824A1
• Filed: 07/28/2011
• Published: 05/23/2013
• Est. Priority Date: 03/01/2011
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Cohen, Scott D., Price, Brian L.
• Granted Patent: US 8,610,712 B2
• Time in Patent Office: Days
• US Class Current: 345/419
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 2207/20096 Interactive definition of curve of interest
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/143 involving probabilistic approaches, e.g. Markov random field [MRF] modelling
• CPC Class Codes: G06T 7/162 involving graph-based methods

Abstract

A method is disclosed. Input data indicates a location within a region of a digital image of a stereo image pair. A boundary of the region and a corresponding boundary of a corresponding region of another digital image are defined based on the input data, content of the digital image, and content of the another digital image. The boundary and the corresponding boundary represent boundaries of a visual feature in the digital image and the other digital image. The defining further includes constructing a probability distribution for the digital image and another probability distribution for the other digital image. The probability distribution for the digital image represents a probability that each pixel of a set of pixels of the digital image is within the region. The other probability distribution represents a probability that each pixel of another set of pixels of the another digital image is within the corresponding region.

Claims

1. A method, comprising,

receiving input data in a display of a set of multiple digital images comprising a stereo image pair, wherein

the input data indicates a location within a region of a digital image of the stereo image pair;

and

defining a boundary of the region and a corresponding boundary of a corresponding region of another digital image, wherein

the boundary and the corresponding boundary represent boundaries of a shared visual feature in the digital image and the another digital image,

the defining is based on the input data, content of the digital image, and content of the another digital image, and

the defining further comprises constructing

a probability distribution for the digital image, wherein the probability distribution represents a probability that each pixel of a set of pixels of the digital image is within the region, and

another probability distribution for the another digital image, wherein the another probability distribution represents a probability that each pixel of another set of pixels of the another digital image is within the corresponding region, and

the probability distribution and the another probability distribution are based on correspondence probabilities between each pixel of the set of pixels and respective ones of the another set of pixels.

2. The method of claim 1, further comprising

displaying the boundary of the region in a display of the digital image, wherein the user input is received in the display of the digital image;

displaying the corresponding boundary of the corresponding region in a display of the another digital image;

receiving other input data, wherein

the other input data indicates a location with respect to which modification of the boundary and the corresponding boundary is requested, and

the other input data is received in the display of the another digital image; and

modifying the boundary and the corresponding boundary in response to the other input data.

3. The method of claim 1, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of color values of respective ones of each pixel of the set of pixels and the respective ones of the another set of pixels.

4. The method of claim 1, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of region labels of respective ones of each pixel of the set of pixels and the respective ones of the another set of pixels, wherein the region labels identify on a binary basis whether a pixel is within or outside of a respective one of the region and the corresponding region.

5. The method of claim 1, wherein

the method further comprises rectifying the digital image and the another digital image;

and

the probability distribution and the another probability distribution are restricted to probability of correspondence for points within a selected disparity along a particular scanline of the digital image and the another digital image.

6. The method of claim 1, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of depth values of ones of each pixel of the set of pixels to the respective ones of the another set of pixels.

7. The method of claim 1, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise normalized correspondence probabilities.

8. A computer-readable storage medium storing program instructions, wherein the program instructions are computer-executable to implement:

receiving input data in a display of a set of multiple digital images comprising a stereo image pair, wherein

the input data indicates a location within a region of a digital image of the stereo image pair; and

defining a boundary of the region and a corresponding boundary of a corresponding region of another digital image, wherein

the boundary and the corresponding boundary represent boundaries of a shared visual feature in the digital image and the another digital image,

the defining is based on the input data, content of the digital image, and content of the another digital image, and

the defining further comprises using

a probability distribution for the digital image, wherein the probability distribution represents a probability that each pixel of a set of pixels of the digital image is within the region, and

another probability distribution for the another digital image, wherein the another probability distribution represents a probability that each pixel of another set of pixels of the another digital image is within the corresponding region, and

the probability distribution and the another probability distribution are based on correspondence probabilities between each pixel of the set of pixels and respective ones of the another set of pixels.

9. The computer-readable storage medium of claim 8, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on disparity values calculated between each pixel of the set of pixels and the respective ones of the another set of pixels, wherein the disparity values represent differences in location between each pixel of the set of pixels and the respective ones of the another set of pixels.

10. The computer-readable storage medium of claim 8, further comprising program instructions computer-executable to implement:

displaying the boundary of the region in a display of the digital image, wherein the user input is received in the display of the digital image;

displaying the corresponding boundary of the corresponding region in a display of the another digital image;

receiving other input data, wherein

the other input data indicates a location with respect to which modification of the boundary and the corresponding boundary is requested, and

the other input data is received in the display of the another digital image; and

modifying the boundary and the corresponding boundary in response to the other input data.

11. The computer-readable storage medium of claim 8, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of color values of respective ones of each pixel of the set of pixels and the respective ones of the another set of pixels.

12. The computer-readable storage medium of claim 8, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of region labels of respective ones of each pixel of the set of pixels and the respective ones of the another set of pixels, wherein the region labels identify on a binary basis whether a pixel is within or outside of a respective one of the region and the corresponding region.

13. The computer-readable storage medium of claim 8, wherein

the computer-readable storage medium further comprises program instructions computer-executable to implement rectifying the digital image and the another digital image;

and

the probability distribution and the another probability distribution are restricted to probability of correspondence for points within a selected disparity along a particular scanline of the digital image and the another digital image.

14. The computer-readable storage medium of claim 8, wherein the correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of depth values of ones of each pixel of the set of pixels to the respective ones of the another set of pixels.

15. A system, comprising:

at least one processor; and

a memory comprising program instructions, wherein the program instructions are executable by the at least one processor to;

receive input data in a display of a set of multiple digital images, wherein the input data indicates a location within a region of a digital image; and

define a boundary of the region and a corresponding boundary of a corresponding region of another digital image, wherein

the boundary and the corresponding boundary represent boundaries of a shared visual feature in the digital image and the another digital image,

the program instructions executable by the at least one processor to define the boundary further comprise program instructions executable by the at least one processor to define the boundary based on the input data, content of the digital image, and content of the another digital image, and

the program instructions executable by the at least one processor to define the boundary further comprise program instructions executable by the at least one processor to construct

a probability distribution for the digital image, wherein the probability distribution represents a probability that each pixel of a set of pixels of the digital image is within the region, and

another probability distribution for the another digital image, wherein the another probability distribution represents a probability that each pixel of another set of pixels of the another digital image is within the corresponding region.

16. The system of claim 15, further comprising program instructions executable by the at least one processor to:

display the boundary of the region in a display of the digital image, wherein the user input is received in the display of the digital image;

display the corresponding boundary of the corresponding region in a display of the another digital image;

receive other input data, wherein

the other input data indicates a location with respect to which modification of the boundary and the corresponding boundary is requested, and

the other input data is received in the display of the another digital image; and

modify the boundary and the corresponding boundary in response to the other input data.

17. The system of claim 15, wherein correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of color values of respective ones of each pixel of the set of pixels and the respective ones of the another set of pixels.

18. The system of claim 15, wherein correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of region labels of respective ones of each pixel of the set of pixels and the respective ones of the another set of pixels, wherein the region labels identify on a binary basis whether a pixel is within or outside of a respective one of the region and the corresponding region.

19. The system of claim 15, wherein

the system further comprises program instructions executable by the at least one processor to implement rectifying the digital image and the another digital image;

and

the probability distribution and the another probability distribution are restricted to probability of correspondence for points within a selected disparity along a particular scanline of the digital image and the another digital image.

20. The system of claim 15, wherein correspondence probabilities between each pixel of the set of pixels and the respective ones of the another set of pixels further comprise correspondence probabilities based on comparison of depth values of ones of each pixel of the set of pixels to the respective ones of the another set of pixels.