Separate range tone mapping for component images in a combined image
Use of separate range tone mapping for combined images can help minimize loss of image information in scenes that have drastically different luminance values, i.e., scenes that have both bright and shadowed regions. Separate range tone mapping is particularly useful for combined images, such as those from spherical camera systems, which may have a higher probability of including luminance variability. The resulting increased bit depth of separate range tone mapping can make the transition between different images that make up a combined image more subtle. Each of a plurality of images that make up a combined image can use a different tone map that is optimized for the particular image data of the image. Multiple tone maps that are applied to overlapping regions of the plurality of images can subsequently be combined to expand the bit depth of the overlapping regions.
1. A system comprising:
a multi-camera array comprising a plurality of cameras having overlapping fields of view and configured to synchronously capture a plurality of images having overlapping regions;
a tone mapping engine configured to produce, for each of the plurality of images, a tone-mapped image having pixels of a first bit depth based on a tone map corresponding to the image, wherein the tone map corresponding to the image is different from at least one other tone map corresponding to another image of the plurality of images; and
an image processing engine configured to combine the tone-mapped images to form a combined image having at least one region corresponding to one or more of the overlapping regions, wherein combining the plurality of tone-mapped images comprises combining the tone maps corresponding to the tone-mapped images to produce a combined tone map such that the at least one region corresponding to the one or more of the overlapping regions has pixels of a second bit depth greater than the first bit depth.
2. The system of claim 1 , wherein the multi-camera array is a spherical camera system.
3. The system of claim 2 , wherein the spherical camera system comprises two cameras, and wherein the plurality of images comprises two images.
4. The system of claim 1 , wherein at least one image of the plurality of images is divided into a plurality of zones, wherein a different tone map is used for at least one of the plurality of zones, the different tone map being different from at least one other tone map used for another zone of the plurality of zones.
5. The system of claim 4 , wherein the at least one image is divided into the plurality of zones based on image content.
6. The system of claim 4 , wherein the at least one image is divided into the plurality of zones based on position.
7. The system of claim 6 , wherein at least one of the plurality of zones is a stitch area that extends past an overlapping region of the at least one image.
8. A method comprising:
accessing a plurality of images having overlapping regions and synchronously captured by a multi-camera array;
producing, for each of the plurality of images, a tone-mapped image having pixels of a first bit depth based on a tone map corresponding to the image, wherein the tone map corresponding to the image is different from at least one other tone map corresponding to another image of the plurality of images; and
combining the tone-mapped images to form a combined image having at least one region corresponding to one or more of the overlapping regions, wherein combining the plurality of tone-mapped images comprises combining the tone maps corresponding to the tone-mapped images to produce a combined tone map such that the at least one region corresponding to the one or more of the overlapping regions has pixels of a second bit depth greater than the first bit depth.
9. The method of claim 8 , wherein the multi-camera array is a spherical camera system.
10. The method of claim 9 , wherein the spherical camera system comprises two cameras, and wherein the plurality of images comprises two images.
11. The method of claim 8 , wherein at least one image of the plurality of images is divided into a plurality of zones, wherein a different tone map is used for at least one of the plurality of zones, the different tone map being different from at least one other tone map used for another zone of the plurality of zones.
12. The method of claim 11 , wherein the at least one image is divided into the plurality of zones based on image content.
13. The method of claim 11 , wherein the at least one image is divided into the plurality of zones based on position.
14. The method of claim 13 , wherein at least one of the plurality of zones is a stitch area that extends past an overlapping region of the at least one image.
15. A non-transitory computer-readable medium comprising instructions that when executed by the processor cause the processor to perform a method comprising:
accessing a plurality of images having overlapping regions and synchronously captured by a multi-camera array;
producing, for each of the plurality of images, a tone-mapped image having pixels of a first bit depth based on a tone map corresponding to the image, wherein the tone map corresponding to the image is different from at least one other tone map corresponding to another image of the plurality of images; and
combining the plurality of tone-mapped images to form a combined image having at least one region corresponding to one or more of the overlapping regions, wherein combining the plurality of tone-mapped images comprises combining the tone maps corresponding to the tone-mapped images to produce a combined tone map such that the at least one region corresponding to the one or more of the overlapping regions has pixels of a second bit depth greater than the first bit depth.
16. The non-transitory computer-readable medium of claim 15 , wherein the multi-camera array is a spherical camera system.
17. The non-transitory computer-readable medium of claim 15 , wherein at least one image of the plurality of images is divided into a plurality of zones, wherein a different tone map is used for at least one of the plurality of zones, the different tone map being different from at least one other tone map used for another zone of the plurality of zones.
18. The non-transitory computer-readable medium of claim 17 , wherein the at least one image is divided into the plurality of zones based on image content.
19. The non-transitory computer-readable medium of claim 17 , wherein the at least one image is divided into the plurality of zones based on position.
20. The non-transitory computer-readable medium of claim 19 , wherein at least one of the plurality of zones is a stitch area that extends past an overlapping region of the at least one image.