Compact array of imaging devices with supplemental imaging unit
A method and system are described. The method includes capturing a set of images from a 2×2 array of cameras, each camera of the array of cameras having an overlapping field of view (FOV) with an adjacent camera of the array of cameras. The method further includes synchronously capturing a supplemental image from a fifth camera, the fifth camera having an at least partially overlapping FOV with every camera of the array of cameras. Supplemental information is extracted by comparing the supplemental image with the set of four images. Portions of the set of images are stitched based in part on the supplemental information to produce a combined stitched image, the combined stitched image having a higher resolution than each image of the set of images.
1. A computer-implemented method for capturing images from a camera array, comprising:
capturing a set of images from a 2×2 array of cameras, each camera of the array of cameras having an overlapping field of view (FOV) with an adjacent camera of the array of cameras;
synchronously capturing a supplemental image from a fifth camera, the fifth camera having an at least partially overlapping FOV with every camera of the array of cameras;
extracting supplemental information by comparing the supplemental image with the set of four images; and
stitching portions of the set of images based in part on the supplemental information to produce a combined stitched image, the combined stitched image having a higher resolution than each image of the set of images.
2. The computer-implemented method of claim 1 , wherein each of the cameras in the array of cameras is rotated by an offset angle from a reference plane, and wherein the combined FOVs of the array of cameras are greater than the individual FOV of each camera.
3. The computer-implemented method of claim 1 , wherein the fifth camera is located between the array of cameras.
4. The computer-implemented method of claim 1 , wherein the fifth camera is located external to the array of cameras.
5. The computer-implemented method of claim 1 , wherein the fifth camera is a depth sensing camera.
6. The computer-implemented method of claim 1 , wherein the supplemental information includes depth information, and wherein stitching portions of the set of images based in part on the supplemental information comprises:
performing image manipulation operations on each of the set of images to correct for a corresponding rotated FOV of each of the camera in the array of cameras;
identifying, using the depth information, objects within portions of the set of images corresponding to the overlapping FOVs;
aligning the set of images such that one or more of the identified objects are aligned; and
stitching the aligned set of images to produce the combined stitched image.
7. The computer-implemented method of claim 6 , wherein the supplemental information additionally includes object tracking information, the object tracking information indicating a location of objects in the shared FOV of the array of cameras, and wherein stitching portions of the set of images based in part on the supplemental information further comprises:
identifying, using the object tracking information, objects within portions of the set of images corresponding to the overlapping FOVs of each camera of the array of cameras.
8. The computer-implemented method of claim 1 , wherein the supplemental information includes depth information, the method further comprising:
generating a three dimensional (3D) reconstruction of a scene captured by the array of cameras using the depth information to produce a 3D model of the scene and using the set of images to texture the 3D model of the scene to produce the 3D reconstruction of the scene.
9. The computer-implemented method of claim 1 , wherein the supplemental information includes depth information, the method further comprising:
generating object tracking information based on the depth information, the object tracking information indicating locations of objects in a combined FOV of the array of cameras; and
maintaining optical focus on an object in the combined FOV of the array of cameras over a plurality of captured images using the object tracking information.
10. The computer-implemented method of claim 1 , wherein the fifth camera captures video for use as a live video feed.
11. A multi-camera system comprising:
a 2×2 array of cameras, each camera of the array of cameras having an overlapping field of view (FOV) with an adjacent camera of the array of cameras, the 2×2 array of cameras configured to capture a set of images;
a fifth camera having an at least partially overlapping FOV with every camera of the array of cameras, the fifth camera configured to capture a supplemental image; and
an image processor configured to: stitch portions of the set of images based at least in part on supplemental information to produce a combined stitched image, the supplemental information extracted based on the supplemental image, the combined stitched image having a higher resolution than each image of the set of images.
12. The system of claim 11 , wherein each of the cameras in the array of cameras is rotated by an offset angle from a reference plane, and wherein the combined FOVs of the array of cameras are greater than the individual FOV of each camera.
13. The system of claim 11 , wherein the fifth camera is located between the array of cameras.
14. The system of claim 11 , wherein the fifth camera is located external to the array of cameras.
15. The system of claim 11 , wherein the fifth camera is a depth sensing camera.
16. The system of claim 11 , wherein the supplemental information includes depth information, and wherein the image processor is further configured to:
perform image manipulation operations on each of the set of images to correct for a corresponding rotated FOV of each of the camera in the array of cameras;
identify, using the depth information, objects within portions of the set of images corresponding to the overlapping FOVs;
align the set of images such that one or more of the identified objects are aligned; and
stitch the aligned set of images to produce the combined stitched image.
17. The system of claim 16 , wherein the supplemental information additionally includes object tracking information, the object tracking information indicating a location of objects in the shared FOV of the array of cameras, and wherein the image processor is further configured to:
identify, using the object tracking information, objects within portions of the set of images corresponding to the overlapping FOVs of each camera of the array of cameras.
18. The system of claim 11 , wherein the supplemental information includes depth information, and wherein the image processor is further configured to:
generate a three dimensional (3D) reconstruction of a scene captured by the array of cameras using the depth information to produce a 3D model of the scene and using the set of images to texture the 3D model of the scene to produce the 3D reconstruction of the scene.
19. The system of claim 11 , wherein the supplemental information includes depth information, and wherein the image processor is further configured to:
generate object tracking information based on the depth information, the object tracking information indicating locations of objects in a combined FOV of the array of cameras; and
maintain optical focus on an object in the combined FOV of the array of cameras over a plurality of captured images using the object tracking information.
20. The system of claim 11 , wherein the fifth camera captures video for use as a live video feed.