Method and device for generating panoramic images

1. A method for generating panoramic images, comprising:

capturing images by at least one camera from a plurality of positions in a circular array;

projecting, by a processor, the captured images into corresponding equirectangular images;

extracting, by the processor, a set of intermediate images from the equirectangular images; and

blending, by the processor, the set of intermediate images to generate panoramic images,

wherein the panoramic images comprise a first panoramic image including scenes for presentation to a right eye of a user and a second panoramic image including scenes for presentation to a left eye of the user.

2. The method of claim 1 , wherein projecting the captured images further comprises:

determining, by the processor, coordinates of pixels in a first coordinate system corresponding to the captured images;

looking up the coordinates in a mapping table;

obtaining corresponding coordinates of pixels in the equirectangular images from the mapping table; and

supplying values of the pixels in the captured images to pixels in the corresponding coordinates of the equirectangular images.

3. The method of claim 1 , wherein extracting a set of intermediate images comprises:

extracting a center portion of each of the equirectangular images as an intermediate image, to obtain the set of intermediate images.

4. The method of claim 1 , wherein extracting a set of intermediate images comprises:

determining from the projected equirectangular images:

a first equirectangular image corresponding to a first image captured at a first position;

a second equirectangular image corresponding to a second image captured at a second position; and

a third equirectangular image corresponding to a third image captured at a third position;

wherein the second position is on a left side of the first position, and the third position is on a right side of the first position in the circular array;

extracting, as a first intermediate image, a first portion of the first equirectangular image that overlaps with the second equirectangular image; and

extracting, as a second intermediate image, a second portion of the first equirectangular image that overlaps with the third equirectangular image.

5. The method of claim 4 , wherein:

the set of intermediate images comprises a first subset of intermediate images and a second subset of intermediate images; and

blending the set of intermediate images comprises:

determining a first overlapping area between two intermediate images in the first subset, and blending the two intermediate images in the first subset, based on the first overlapping area, to obtain the first panoramic image; and

determining a second overlapping area between two intermediate images in the second subset, and blending the two intermediate images in the second subset, based on the second overlapping area, to obtain the second panoramic image.

6. The method of claim 1 , wherein blending the set of the intermediate images comprises:

determining an overlapping area between two intermediate images in the set of the intermediate images; and

blending the two intermediate images based on the overlapping area.

7. The method of claim 1 , wherein capturing images from positions in a circular array comprises:

capturing the images by sixteen cameras respectively disposed at the positions, facing radially outward.

8. The method of claim 1 , wherein capturing images from positions in a circular array comprises:

capturing a first image by one camera at a first position; and

rotating the camera by an angle to a second position to capture a second image.

9. The method of claim 8 , wherein the angle is 22.5 degrees.

10. A device for generating panoramic images, comprising:

a camera rig;

at least one camera, mounted on the camera rig and configured to capture images from a plurality of positions in a circular array;

a memory, configured to store the captured images and instructions; and

a processor, configured to execute the instructions to:

project the captured images into corresponding equirectangular images;

extract a set of intermediate images from the equirectangular images; and

blend the set of intermediate images to generate panoramic images,

wherein the panoramic images comprise a first panoramic image including scenes for presentation to a right eye of a user and a second panoramic image including scenes for presentation to a left eye of the user.

11. The device of claim 10 , wherein the processor further executes the instructions to project the captured images into corresponding equirectangular images by:

determining coordinates of pixels in the captured images in a first coordinate system;

looking up the coordinates in a mapping table;

obtaining corresponding coordinates of pixels in the equirectangular images in a second coordinate system from the mapping table; and

supplying values of the pixels in the captured images to pixels in the corresponding coordinates of the equirectangular images.

12. The device of claim 10 , wherein the processor further executes the instructions to extract a set of intermediate images by:

extracting a center portion of each of the equirectangular images as intermediate images, to obtain the set of intermediate images.

13. The device of claim 10 , wherein the processor further executes the instructions to extract a set of intermediate images by:

determining from the projected equirectangular images:

a first equirectangular image corresponding to a first image captured at a first position;

a second equirectangular image corresponding to a second image captured at a second position; and

a third equirectangular image corresponding to a third image captured at a third position;

wherein the second position is on a left side of the first position, and the third position is on a right side of the first position in the circular array;

extracting, as a first intermediate image, a first portion of the first equirectangular image that overlaps with the second equirectangular image; and

extracting, as a second intermediate image, a second portion of the first equirectangular image that overlaps with the third equirectangular image.

14. The device of claim 13 , wherein:

the set of intermediate images comprises a first subset of intermediate images and a second subset of intermediate images; and

the processor further executes the instructions to blend the set of intermediate images by:

determining a first overlapping area between two intermediate images in the first subset of intermediate images, and blending the two intermediate images based on the first overlapping area to obtain the first panoramic image; and

determining a second overlapping area between two intermediate images in the second subset of intermediate images, and blending the two intermediate images based on the second overlapping area to obtain the second panoramic image.

15. The device of claim 10 , wherein the processor further executes the instructions to blend the set of intermediate images by:

determining an overlapping area between two intermediate images in the set of intermediate images; and

blending the two intermediate images based on the overlapping area.

16. The device of claim 10 , wherein the device includes sixteen cameras respectively disposed at the positions, facing radially outward.

17. The device of claim 10 , further comprising a rotation unit, wherein the device includes one camera, and captures a first image by the camera at a first position, and rotates, via the rotation unit, the camera by an angle to a second position to capture a second image.

18. The device of claim 17 , wherein the angle is 22.5 degrees.