Material capture using imaging

1. A computer-implemented method for material capture of a surface, comprising:

receiving a plurality of images, each of the plurality of images depicting a material surface;

aligning the plurality of images based on alignment of corresponding pixels associated with each of the plurality of images;

determining reflectance information for at least a portion of the corresponding pixels associated with each of the plurality of images;

selecting a set of reference materials that represent properties of the material surface based on the determined reflectance information, the selected reference materials representing properties of the material surface;

generating a material model using the set of reference materials in conjunction with the determined reflectance information; and

outputting the material model of the material surface.

2. The computer-implemented method for material capture of a surface of claim 1 , wherein determining the reflectance information includes finding lighting and view directions for at least one pixel of the corresponding pixels.

3. The computer-implemented method for material capture of a surface of claim 2 , wherein a reference grid is used for aligning the plurality of images.

4. The computer-implemented method for material capture of a surface of claim 1 , further comprising:

determining a surface normal estimate at a selected pixel;

determining a material coefficient abundance estimate at the selected pixel;

iteratively updating the surface normal estimate and updating the material coefficient abundance estimate at the selected pixel until error is reduced below a predetermined threshold.

5. The computer-implemented method for material capture of a surface of claim 4 , wherein the surface normal estimate is determined utilizing a local neighborhood search of a grid based on elevation and azimuth angles for the set of reference materials.

6. The computer-implemented method for material capture of a surface of claim 4 , wherein the material coefficient abundance estimate is determined utilizing a non-negative least squares approach based on inputting the surface normal estimate.

7. The computer-implemented method for material capture of a surface of claim 1 , further comprising:

determining a surface normal estimate at a selected pixel;

determining a material coefficient abundance estimate at the selected pixel;

updating the surface normal estimate and updating the material coefficient abundance estimate at the selected pixel in parallel until error is reduced below a predetermined threshold.

8. The computer-implemented method for material capture of a surface of claim 1 , wherein selecting the set of reference materials includes generating a sparse response curve for a library of reference materials, wherein reference materials with a response on the sparse response curve are selected as the set of reference materials.

9. The computer-implemented method for material capture of a surface of claim 1 , wherein a user can apply the material model of the material surface to a surface of a computer-generated object during rendering of the computer-generated object.

10. One or more non-transitory computer-readable storage media having instructions stored thereon, which, when executed by one or more processors of a computing device, cause the computing device to:

receive a plurality of images, the plurality of images depicting a material surface;

calibrate the plurality of images, wherein calibration utilizes a grid depicted in the plurality of images to align the plurality of images and determine reflectance information for at least a portion of the aligned plurality of images;

select a set of reference materials using the calibrated plurality of images;

determine material properties for the material surface at a selected pixel using the set of reference materials; and

output a material model of the material surface based on the determined material properties.

11. The one or more non-transitory computer-readable storage media of claim 10 , wherein the reflectance information includes lighting and view directions for at least one pixel of the plurality of images.

12. The one or more non-transitory computer-readable storage media of claim 10 , wherein instructions further cause the computing device to:

determine a surface normal estimate at the selected pixel;

determine a material coefficient abundance estimate at the selected pixel;

iteratively update the surface normal estimate and update the material coefficient abundance estimate at the selected pixel until error is reduced below a predetermined threshold.

13. The one or more non-transitory computer-readable storage media of claim 12 , wherein the surface normal estimate is determined utilizing a local neighborhood search of a grid based on elevation and azimuth angles for the set of reference materials.

14. The one or more non-transitory computer-readable storage media of claim 12 , wherein the material coefficient abundance estimate is determined utilizing a non-negative least squares approach based on inputting the surface normal estimate.

15. The one or more non-transitory computer-readable storage media of claim 10 , wherein instructions further cause the computing device to:

determine a surface normal estimate at the selected pixel;

determine a material coefficient abundance estimate at the selected pixel;

update the surface normal estimate and update the material coefficient abundance estimate at the selected pixel in parallel until error is reduced below a predetermined threshold.

16. The one or more non-transitory computer-readable storage media of claim 10 , wherein determining global material estimation includes generating a sparse response curve for a library of reference materials, wherein reference materials with a response on the sparse response curve are selected as the set of reference materials.

17. A computing system comprising:

one or more processors; and

one or more non-transitory computer-readable storage media, coupled with the one or more processors, having instructions stored thereon, which, when executed by the one or more processors, cause the computing system to provide:

means for capturing a plurality of images depicting a material surface;

means for calibrating the plurality of images;

means for selecting a set of reference materials;

means for determining material properties for the material surface at a selected pixel using the set of reference materials;

means for outputting a material model of the material surface based on the determined material properties for the material surface; and

means for applying the material model of the material surface to a surface of a computer-generated object.

18. The computing system of claim 17 further comprising:

means for determining a surface normal estimate at the selected pixel;

means for determining a material coefficient abundance estimate at the selected pixel;

means for updating the surface normal estimate and updating the material coefficient abundance estimate at the selected pixel until error is reduced below a predetermined threshold.

19. The computing system of claim 17 , wherein determining global material estimation includes generating a sparse response curve for a library of reference materials, wherein reference materials with a response on the sparse response curve are selected as the set of reference materials.

20. The computing system of claim 17 , wherein the material coefficient abundance estimate is determined utilizing a non-negative least squares approach based on inputting the surface normal estimate.