Expanding a digital representation of a physical plane

• US 9,595,125 B2
• Filed: 08/28/2014
• Issued: 03/14/2017
• Est. Priority Date: 08/30/2013
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Reisner-Kollmann, Irene, Pan, Qi, Siklossy, Istvan, Gervautz, Michael, Nielsen, Per Orum, Lawrence Ashok Inigo, Roy
• Primary Examiner(s): NGUYEN, ANH TUAN V
• Application Number:
  US14/472,288
• Publication Number:
  US 20150062166A1
• Time in Patent Office: 929 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06T 19/006 Mixed reality object pose determination, tracking or camera calibration for mixed reality G06T7/00
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80

Abstract

Techniques are presented for expanding a digital representation of a physical plane from a physical scene. In some aspects, a method may include determining an orientation and an initial portion of a physical plane in the scene, and subdividing a rectified image for the scene into a plurality of grid cells. For the grid cells, an image signature may be generated. A grid cell contiguous to the obtained initial portion of the plane is determined to include part of the plane. An iterative process may be performed for each neighboring grid cell from the grid cell contiguous to at least part of the obtained initial portion, determining whether the neighboring grid cell is to be included as part of the plane if the image signature of said neighboring grid cell is similar to the image signature of a grid cell already determined to be included as part of the plane.

Claims

1. A method for constructing a digital representation of a physical plane from a physical scene, the method comprising:

determining an orientation and an initial portion of a physical plane in the physical scene in three dimensional space;

generating a three dimensional rectified image, wherein the rectified image contains a fronto-parallel view of at least part of the physical plane and wherein the rectified image is generated by matching reference points in a plurality of two dimensional images of the physical plane captured from different locations with different distances to the reference points in the physical scene;

subdividing the rectified image into a plurality of grid cells;

for each of the plurality of grid cells, generating an image signature based on at least one image characteristic associated with said grid cell, wherein the at least one image characteristic comprises three dimensional information derived from the reference points;

selecting one or more initial grid cells contiguous to at least part of the initial portion of the physical plane to include as part of the digital representation of the physical plane;

comparing, as a first comparison, the image signature for each neighboring grid cell contiguous to the one or more initial grid cells to the image signature for one or more grid cells currently included in the digital representation of the physical plane; and

adding each grid cell that meets a predetermined threshold for the first comparison to the digital representation of the physical plane.

2. The method of claim 1, further comprising:

comparing, as a second comparison, the image signature for each neighboring grid cell contiguous to a grid cell already included as part of the digital representation of the physical plane to the image signature for the contiguous grid cell; and

adding each grid cell that meets the predetermined threshold for the second comparison to the digital representation of the physical plane.

3. The method of claim 1, wherein the initial portion of the physical plane is based on a known object positioned on the physical plane.

4. The method of claim 1, wherein the initial portion of the physical plane is based on fitting the physical plane to a point cloud model.

5. The method of claim 1, wherein the image signature comprises at least one histogram indicative of image information associated with the grid cell.

6. The method of claim 5, wherein the at least one histogram comprises a statistical representation of an amount and type of colors present in the grid cell.

7. The method of claim 5, wherein the at least one histogram comprises a statistical representation of a gradient orientation of the grid cell.

8. The method of claim 5, wherein the at least one histogram comprises a statistical representation of an amount and type of gray scale shades representative of colors present in the grid cell.

9. The method of claim 1, further comprising, for each neighboring grid cell determined to not be included as part of the digital representation of the physical plane:

generating a plurality of secondary grid cells, wherein a position of each of the plurality of secondary grid cells is based on an offset from a position of said neighboring grid cell, wherein the offset is within a width and a length of said neighboring grid cell; and

for each of the plurality of secondary grid cells;

generating an image signature based on at least one image characteristic associated with the secondary grid cell; and

determining whether said secondary grid cell is to be additionally included as part of the digital representation of the physical plane if the image signature of said secondary grid cell is similar to the image signature of a grid cell already determined to be included as part of the digital representation of the physical plane, wherein the similarity is based on at least one predetermined threshold criterion.

10. An apparatus for constructing a digital representation of a physical plane from a physical scene, comprising:

memory;

one or more processors configured to;

determine an orientation and an initial portion of a physical plane in the physical scene in three dimensional space;

receive a plurality of two dimensional images of the physical plane;

generate a three dimensional rectified image, wherein the rectified image contains a fronto-parallel view of at least part of the physical plane and wherein the rectified image is generated by matching reference points in the plurality of two dimensional images of the physical plane captured from different locations with different distances to the reference points in the physical scene;

subdivide the rectified image into a plurality of grid cells;

for each of the plurality of grid cells, generate an image signature based on at least one image characteristic associated with said grid cell, wherein the at least one image characteristic comprises three dimensional information derived from the reference points;

select one or more initial grid cells contiguous to at least part of the initial portion of the physical plane to include as part of the digital representation of the physical plane;

compare, as a first comparison, the image signature for each neighboring grid cell contiguous to the one or more initial grid cells to the image signature for one or more grid cells currently included in the digital representation of the physical plane; and

add each grid cell that meets a predetermined threshold for the first comparison to the digital representation of the physical plane.

11. The apparatus of claim 10, further comprising:

comparing, as a second comparison, the image signature for each neighboring grid cell contiguous to a grid cell already included as part of the digital representation of the physical plane to the image signature for the contiguous grid cell; and

adding each grid cell that meets the predetermined threshold for the second comparison to the digital representation of the physical plane.

12. The apparatus of claim 10, wherein the initial portion of the physical plane is based on a known object positioned on the physical plane.

13. The apparatus of claim 10, wherein the initial portion of the physical plane is based on fitting the physical plane to a point cloud model.

14. The apparatus of claim 10, wherein the image signature comprises at least one histogram indicative of image information associated with the grid cell.

15. The apparatus of claim 14, wherein the at least one histogram comprises a statistical representation of an amount and type of colors present in the grid cell.

16. The apparatus of claim 14, wherein the at least one histogram comprises a statistical representation of a gradient orientation of the grid cell.

17. The apparatus of claim 14, wherein the at least one histogram comprises a statistical representation of an amount and type of gray scale shades representative of colors present in the grid cell.

18. The apparatus of claim 10, for each neighboring grid cell determined to not be included as part of the digital representation of the physical plane, the one or more processors further configured to:

generate a plurality of secondary grid cells, wherein a position of each of the plurality of secondary grid cells is based on an offset from a position of said neighboring grid cell, wherein the offset is within a width and a length of said neighboring grid cell; and

for each of the plurality of secondary grid cells;

generate an image signature based on at least one image characteristic associated with the secondary grid cell; and

determine whether said secondary grid cell is to be additionally included as part of the digital representation of the physical plane if the image signature of said secondary grid cell is similar to the image signature of a grid cell already determined to be included as part of the digital representation of the physical plane, wherein the similarity is based on at least one predetermined threshold criterion.

19. A non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium comprises instructions executable by a processor, the instructions comprising instructions to:

determine an orientation and an initial portion of a physical plane in the physical scene in three dimensional space;

receive a plurality of two dimensional images of the physical plane;

generate a three dimensional rectified image, wherein the rectified image contains a fronto-parallel view of at least part of the physical plane and wherein the rectified image is generated by matching reference points in the plurality of two dimensional images of the physical plane captured from different locations with different distances to the reference points in the physical scene;

subdivide the rectified image into a plurality of grid cells;

for each of the plurality of grid cells, generate an image signature based on at least one image characteristic associated with said grid cell, wherein the at least one image characteristic comprises three dimensional information derived from the reference points;

select one or more initial grid cells contiguous to at least part of the initial portion of the physical plane to include as part of the digital representation of the physical plane;

compare, as a first comparison, the image signature for each neighboring grid cell contiguous to the one or more initial grid cells to the image signature for one or more grid cells currently included in the digital representation of the physical plane; and

add each grid cell that meets a predetermined threshold for the first comparison to the digital representation of the physical plane.

20. The non-transitory computer-readable storage medium of claim 19, further comprising:

comparing, as a second comparison, the image signature for each neighboring grid cell contiguous to a grid cell already included as part of the digital representation of the physical plane to the image signature for the contiguous grid cell; and

adding each grid cell that meets the predetermined threshold for the second comparison to the digital representation of the physical plane.

21. The non-transitory computer-readable storage medium of claim 19, wherein the image signature comprises at least one histogram indicative of image information associated with the grid cell.

22. The non-transitory computer-readable storage medium of claim of claim 19, wherein for each neighboring grid cell determined to not be included as part of the digital representation of the physical plane, the one or more processors further configured to:

generate a plurality of secondary grid cells, wherein a position of each of the plurality of secondary grid cells is based on an offset from a position of said neighboring grid cell, wherein the offset is within a width and a length of said neighboring grid cell; and

for each of the plurality of secondary grid cells;

generate an image signature based on at least one image characteristic associated with the secondary grid cell; and

determine whether said secondary grid cell is to be additionally included as part of the digital representation of the physical plane if the image signature of said secondary grid cell is similar to the image signature of a grid cell already determined to be included as part of the digital representation of the physical plane, wherein the similarity is based on at least one predetermined threshold criterion.

23. The apparatus for constructing a digital representation of a physical plane from a physical scene, the apparatus comprising:

means for determining an orientation and an initial portion of a physical plane of the physical scene in three dimensional space;

means for generating a three dimensional rectified image, wherein the rectified image contains a fronto-parallel view of at least part of the physical plane and wherein the rectified image is generated by matching reference points in a plurality of two dimensional images of the physical plane captured from different locations with different distances to the reference points in the physical scene;

means for subdividing the rectified image into a plurality of grid cells;

for each of the plurality of grid cells, means for generating an image signature based on at least one image characteristic associated with said grid cell, wherein the at least one image characteristic comprises three dimensional information derived from the reference points;

means for selecting one or more initial grid cells contiguous to at least part of the initial portion of the physical plane to include as part of the digital representation of the physical plane;

means for comparing, as a first comparison, the image signature for each neighboring grid cell contiguous to the one or more initial grid cells to the image signature for one or more grid cells currently included in the digital representation of the physical plane; and

means for adding each grid cell that meets a predetermined threshold for the first comparison to the digital representation of the physical plane.

24. The apparatus of claim 23, further comprising:

means for comparing, as a second comparison, the image signature for each neighboring grid cell contiguous to a grid cell already included as part of the digital representation of the physical plane to the image signature for the contiguous grid cell; and

means for adding each grid cell that meets the predetermined threshold for the second comparison to the digital representation of the physical plane.

25. The apparatus of claim 23, wherein the image signature comprises at least one histogram indicative of image information associated with the grid cell.

26. The apparatus of claim 23, further comprising, for each neighboring grid cell determined to not be included as part of the digital representation of the physical plane:

means for generating a plurality of secondary grid cells, wherein a position of each of the plurality of secondary grid cells is based on an offset from a position of said neighboring grid cell, wherein the offset is within a width and a length of said neighboring grid cell; and

for each of the plurality of secondary grid cells;

means for generating an image signature based on at least one image characteristic associated with the said secondary grid cell; and

means for determining whether said secondary grid cell is to be additionally included as part of the digital representation of the physical plane if the image signature of said secondary grid cell is similar to the image signature of a grid cell already determined to be included as part of the digital representation of the physical plane, wherein the similarity is based on at least one predetermined threshold criterion.