Three hundred sixty degree video stitching

1. An electronic device configured to render a three hundred and sixty (360) degree image, the electronic device comprising:

a communication unit configured to receive a first image, a second image, and metadata from a camera; and

a processor configured to render the 360 degree image by:

transforming the first image and the second image to compensate for a light fall-off;

unwarping the transformed first image and second image;

aligning the unwarped first image and second image; and

blending an overlapping portion of the aligned unwarped first image and second image to render the 360 degree image.

2. The electronic device of claim 1 , wherein the communication unit is configured to transmit the rendered 360 degree image to a user equipment (UE).

3. The electronic device of claim 1 , wherein transforming the first image and the second image to compensate for a light fall-off is based on a polynomial equation included in the metadata.

4. The electronic device of claim 1 , wherein aligning the unwarped first image and second image comprises:

transforming the unwarped first image and second image to compensate for a mechanical misalignment of the camera; and

transforming the unwarped first image and second image based on at least one of an object or a scene in the unwarped first image and second image.

5. The electronic device of claim 4 , wherein transforming the unwarped first image and second image to compensate for a mechanical misalignment is based on an affine matrix included in the metadata.

6. The electronic device of claim 4 , wherein transforming the unwarped first image and second image based on at least one of an object or a scene in the unwarped first image and second image comprises:

determining a first template and a reference to match in at least one boundary shared between the unwarped first image and second image;

matching the first template to return a correlation matrix;

determining a displacement between the first template and the reference;

determining a second template based on the displacement;

determining an affine matrix based on the first template and the second template; and

warping at least one of the first image or the second image based on the affine matrix.

7. The electronic device of claim 6 , wherein matching the first template is based on a normalized cross-correlation.

8. The electronic device of claim 6 , wherein the affine matrix is determined based on a plurality of vertices in the first template and the second template.

9. A method for rendering a three hundred and sixty (360) degree image in an electronic device having a processor, the method comprising:

receiving a first image, a second image, and metadata from a camera;

transforming, by the processor, the first image and the second image to compensate for a light fall-off;

unwarping, by the processor, the transformed first image and second image;

aligning, by the processor, the unwarped first image and second image; and

blending, by the processor, an overlapping portion of the aligned unwarped first image and second image to render the 360 degree image.

10. The method of claim 9 , further comprising transmitting the rendered 360 degree image to a user equipment (UE).

11. The method of claim 9 , wherein transforming the first image and the second image to compensate for a light fall-off is based on a polynomial equation included in the metadata.

12. The method of claim 9 , wherein aligning the unwarped first image and second image comprises:

transforming the unwarped first image and second image to compensate for a mechanical misalignment of the camera; and

transforming the unwarped first image and second image based on at least one of an object or a scene in the unwarped first image and second image.

13. The method of claim 12 , wherein transforming the unwarped first image and second image to compensate for a mechanical misalignment is based on an affine matrix included in the metadata.

14. The method of claim 12 , wherein transforming the unwarped first image and second image based on at least one of an object or a scene in the unwarped first image and second image comprises:

determining a first template and a reference to match in at least one boundary shared between the unwarped first image and second image;

matching the first template to return a correlation matrix;

determining a displacement between the first template and the reference;

determining a second template based on the displacement;

determining an affine matrix based on the first template and the second template; and

warping at least one of the first image or the second image based on the affine matrix.

15. The method of claim 14 , wherein matching the template is based on a normalized cross-correlation.

16. The method of claim 14 , wherein the affine matrix is determined based on a plurality of vertices in the first template and the second template.

17. A non-transitory computer readable medium embodying a computer program, the computer program comprising computer readable program code that when executed causes at least one processing device to:

receive a first image, a second image, and metadata from a camera; and

transform the first image and the second image to compensate for a light fall-off;

unwarp the transformed first image and second image;

align the unwarped first image and second image; and

blend an overlapping portion of the aligned unwarped first image and second image to render a 360 degree image.

18. The non-transitory computer readable medium of claim 17 , wherein the at least one processing device transforms the first image and the second image to compensate for a light fall-off based on a polynomial equation included in the metadata.

19. The non-transitory computer readable medium of claim 17 , wherein the at least one processing device aligns the unwarped first image and second image by:

transforming the unwarped first image and second image to compensate for a mechanical misalignment of the camera; and

transforming the unwarped first image and second image based on at least one of an object or a scene in the unwarped first image and second image.

20. The non-transitory computer readable medium of claim 19 , wherein the at least one processing device transforms the unwarped first image and second image based on at least one of an object or a scene in the unwarped first image and second image by:

determining a first template and a reference to match in at least one boundary shared between the unwarped first image and second image;

matching the first template to return a correlation matrix;

determining a displacement between the first template and the reference;

determining a second template based on the displacement;

determining an affine matrix based on the first template and the second template; and

warping at least one of the first image or the second image based on the affine matrix.