Primary and auxiliary image capture devices for image processing and related methods
First Claim
1. A method for using primary and auxiliary image capture devices, each including an image sensor and a lens, to perform image processing, the method comprising:
- using the primary image capture device to capture a first image of a scene, the first image having a first quality characteristic;
using the auxiliary image capture device to capture a second image of the scene, the second image having a second quality characteristic, and the second quality characteristic being of lower quality than the first quality characteristic;
adjusting at least one parameter of one of the captured images to create at least one adjusted image for one of approximating and matching a quality characteristic of the other image;
forming an adjusted image pair that includes two images from a first group of images including the adjusted and captured images;
extracting image features using one or more pixels from the images in the adjusted image pair;
matching extracted image features between the images in the adjusted image pair;
computing a transformation to align at least one of the matched image features between the images in the adjusted image pair;
applying the transformation to at least one of the images in the adjusted image pair to align the images in the adjusted image pair;
forming a transformed image pair by selecting two images from a second group of images including the transformed, adjusted, and captured images;
and utilizing the transformed image pair tocreate another image segment utilizing lower quality pixels obtained directly from their corresponding locations from the transformed image pair and higher quality pixels obtained byutilizing disparity information and higher quality pixels from the transformed image pair to replace corresponding pixels on a lower quality image in response to determining that a corresponding image segment of the lower quality image does not meet predefined quality criteria, wherein the replaced corresponding pixels are non-occluded pixels.
1 Assignment
0 Petitions
Accused Products
Abstract
Disclosed herein are primary and auxiliary image capture devices for image processing and related methods. According to an aspect, a method may include using primary and auxiliary image capture devices to perform image processing. The method may include using the primary image capture device to capture a first image of a scene, the first image having a first quality characteristic. Further, the method may include using the auxiliary image capture device to capture a second image of the scene. The second image may have a second quality characteristic. The second quality characteristic may be of lower quality than the first quality characteristic. The method may also include adjusting at least one parameter of one of the captured images to create a plurality of adjusted images for one of approximating and matching the first quality characteristic. Further, the method may include utilizing the adjusted images for image processing.
-
Citations
26 Claims
-
1. A method for using primary and auxiliary image capture devices, each including an image sensor and a lens, to perform image processing, the method comprising:
-
using the primary image capture device to capture a first image of a scene, the first image having a first quality characteristic; using the auxiliary image capture device to capture a second image of the scene, the second image having a second quality characteristic, and the second quality characteristic being of lower quality than the first quality characteristic; adjusting at least one parameter of one of the captured images to create at least one adjusted image for one of approximating and matching a quality characteristic of the other image; forming an adjusted image pair that includes two images from a first group of images including the adjusted and captured images; extracting image features using one or more pixels from the images in the adjusted image pair; matching extracted image features between the images in the adjusted image pair; computing a transformation to align at least one of the matched image features between the images in the adjusted image pair; applying the transformation to at least one of the images in the adjusted image pair to align the images in the adjusted image pair; forming a transformed image pair by selecting two images from a second group of images including the transformed, adjusted, and captured images; and utilizing the transformed image pair to create another image segment utilizing lower quality pixels obtained directly from their corresponding locations from the transformed image pair and higher quality pixels obtained by utilizing disparity information and higher quality pixels from the transformed image pair to replace corresponding pixels on a lower quality image in response to determining that a corresponding image segment of the lower quality image does not meet predefined quality criteria, wherein the replaced corresponding pixels are non-occluded pixels. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)
-
Specification