Method, apparatus and computer program product for generating super-resolved images

1. A method comprising:

facilitating receipt of a light-field image;

determining one or more depth levels in the light-field image;

generating a plurality of images from the light-field image;

determining one or more registration matrices corresponding to the one or more depth levels between an image and one or more remaining images of the plurality of images, wherein the determination of the one or more registration matrices comprises computing the one or more registration matrices based on a plurality of feature points in selected regions of the image and the image of the one or more remaining images; and

performing a super-resolution of the image and the one or more remaining images based on the one or more registration matrices to generate a super-resolved image of the image.

2. The method as claimed in claim 1 , wherein determining a registration matrix of the one or more registration matrices corresponding to a depth level of the one or more depth levels between the image and an image of the one or more remaining images comprises:

selecting a region associated with the depth level in the image and a region associated with the depth level in the image of the one or more remaining images;

determining a plurality of feature points in the selected region in the image and a plurality of feature points in the selected region in the image of the one or more remaining images; and

computing the registration matrix based on correspondence between the plurality of feature points in selected regions of the image and the image of the one or more remaining images.

3. The method as claimed in claim 1 , wherein performing the super-resolution of the image comprises:

performing super-resolutions of region associated with one or more depth levels in the image and regions associated with the one or more depth levels in the one or more remaining images based on the registration matrices corresponding to the one or more depth levels; and

generating the super-resolved image based on the super-resolutions of the regions associated with the one or more depth levels of the image with corresponding regions in the one or more remaining images.

4. The method as claimed in claim 1 , further comprising determining a depth map associated with the one or more depth levels in the light-field image.

5. The method as claimed in claim 1 , wherein performing the super-resolution of the image and the one or more remaining images includes performing a non-uniform interpolation method.

6. The method as claimed in claim 1 , wherein performing the super-resolution of the image and the one or more remaining images includes performing a back projection method.

7. The method as claimed in claim 1 , wherein the super-resolved image has higher resolution than that of the image.

8. An apparatus comprising:

at least one processor; and

at least one memory comprising computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

facilitate receipt of a light-field image;

determine one or more depth levels in the light-field image;

generate a plurality of images from the light-field image;

determine one or more registration matrices corresponding to the one or more depth levels between an image and one or more remaining images of the plurality of images, wherein the determination of the one or more registration matrices comprises computing the one or more registration matrices based on a plurality of feature points in selected regions of the image and the image of the one or more remaining images; and

perform a super-resolution of the image and the one or more remaining images based on the one or more registration matrices to generate a super-resolved image of the image.

9. The apparatus as claimed in claim 8 , wherein for determining a registration matrix of the one or more registration matrices corresponding to a depth level of the one or more depth levels between the image and an image of the one or more remaining images, the apparatus is further caused, at least in part to:

select a region associated with the depth level in the image and a region associated with the depth level in the image of the one or more remaining images;

determine a plurality of feature points in the selected region in the image and a plurality of feature points in the selected region in the image of the one or more remaining images; and

compute the registration matrix based on correspondence between the plurality of feature points in selected regions of the image and the image of the one or more remaining images.

10. The apparatus as claimed in claim 8 , wherein for performing the super-resolution of the image, the apparatus is further caused, at least in part to:

perform super-resolutions of region associated with one or more depth levels in the image and regions associated with the one or more depth levels in the one or more remaining images based on the registration matrices corresponding to the one or more depth levels; and

generate the super-resolved image based on the super-resolutions of the regions associated with the one or more depth levels of the image with corresponding regions in the one or more remaining images.

11. The apparatus as claimed in claim 8 , wherein the apparatus is further caused, at least in part to determine a depth map associated with the one or more depth levels in the light-field image.

12. The apparatus as claimed in claim 8 , wherein for performing the super-resolution of the image and the one or more remaining images, the apparatus is further caused, at least in part to perform a non-uniform interpolation method.

13. The apparatus as claimed in claim 8 , wherein for performing the super-resolution of the image and the one or more remaining images, the apparatus is further caused, at least in part to, perform a back projection method.

14. The apparatus as claimed in claim 8 , wherein the apparatus comprises at least one image sensor configured to capture the light-field image.

15. A computer program product comprising at least one computer-readable storage medium, the computer-readable storage medium comprising a set of instructions, which, when executed by one or more processors, cause an apparatus to at least perform:

facilitate receipt of a light-field image;

determine one or more depth levels in the light-field image;

generate a plurality of images from the light-field image;

determine one or more registration matrices corresponding to the one or more depth levels between an image and one or more remaining images of the plurality of images, wherein the determination of the one or more registration matrices comprises computing the one or more registration matrices based on a plurality of feature points in selected regions of the image and the image of the one or more remaining images; and

perform a super-resolution of the image and the one or more remaining images based on the one or more registration matrices to generate a super-resolved image of the image.

16. The computer program product as claimed in claim 15 , wherein for determining a registration matrix of the one or more registration matrices corresponding to a depth level of the one or more depth levels between the image and an image of the one or more remaining images, the apparatus is further caused, at least in part to:

select a region associated with the depth level in the image and a region associated with the depth level in the image of the one or more remaining images;

determine a plurality of feature points in the selected region in the image and a plurality of feature points in the selected region in the image of the one or more remaining images; and

compute the registration matrix based on correspondence between the plurality of feature points in selected regions of the image and the image of the one or more remaining images.

17. The computer program product as claimed in claim 15 , wherein for performing the super-resolution of the image, the apparatus is further caused, at least in part to:

perform super-resolutions of region associated with one or more depth levels in the image and regions associated with the one or more depth levels in the one or more remaining images based on the registration matrices corresponding to the one or more depth levels; and

generate the super-resolved image based on the super-resolutions of the regions associated with the one or more depth levels of the image with corresponding regions in the one or more remaining images.

18. The computer program product as claimed in claim 15 , wherein the apparatus is further caused, at least in part to determine a depth map associated with the one or more depth levels in the light-field image.

19. The computer program product as claimed in claim 15 , wherein for performing the super-resolution of the image and the one or more remaining images, the apparatus is further caused, at least in part to perform a non-uniform interpolation method.

20. The computer program product as claimed in claim 15 , wherein for performing the super-resolution of the image and the one or more remaining images, the apparatus is further caused, at least in part to, perform a back projection method.