Multi-aperture depth map using blur kernels and edges
First Claim
Patent Images
1. A computer-implemented method for processing blurred image data, comprising:
- selecting a plurality of first windows from first image data associated with a first image of an object, the first image captured using a first imaging system;
selecting a corresponding plurality of second windows from second image data associated with a second image of the object wherein corresponding first and second windows contain a same edge in the object, the second image captured using a second imaging system, wherein a comparison of blurring by the first imaging system and blurring by the second imaging system varies as a function of object depth;
for pairs of corresponding first and second windows, estimating the comparison of blurring by the first and second imaging systems based on blurring of the same edge in corresponding first and second windows, wherein estimating the comparison of blurring by the first and second imaging systems comprises;
binarizing the same edge in corresponding first and second windows, anddetermining a blur kernel that approximates blurring of the same binarized edge in corresponding first and second windows, wherein different blur kernels correspond to different object depths; and
generating depth information for the object based on said estimated comparisons comprises selecting the object depth that corresponds to the determined blur kernel.
1 Assignment
0 Petitions
Accused Products
Abstract
The present disclosure overcomes the limitations of the prior art by using blurring of edges. For example, a first image may contain an edge and a second image may contain the same edge as the first image. The two images may be captured by imaging systems with blur characteristics that vary differently as a function of object depth. For example, a dual-aperture system may simultaneously capture a faster f-number visible image and a slower f-number infrared image. Depth information may be generated by comparing blurring of the same edge in the two images.
126 Citations
18 Claims
-
1. A computer-implemented method for processing blurred image data, comprising:
-
selecting a plurality of first windows from first image data associated with a first image of an object, the first image captured using a first imaging system; selecting a corresponding plurality of second windows from second image data associated with a second image of the object wherein corresponding first and second windows contain a same edge in the object, the second image captured using a second imaging system, wherein a comparison of blurring by the first imaging system and blurring by the second imaging system varies as a function of object depth; for pairs of corresponding first and second windows, estimating the comparison of blurring by the first and second imaging systems based on blurring of the same edge in corresponding first and second windows, wherein estimating the comparison of blurring by the first and second imaging systems comprises; binarizing the same edge in corresponding first and second windows, and determining a blur kernel that approximates blurring of the same binarized edge in corresponding first and second windows, wherein different blur kernels correspond to different object depths; and generating depth information for the object based on said estimated comparisons comprises selecting the object depth that corresponds to the determined blur kernel. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A non-transitory computer-readable storage medium storing executable computer program instructions for processing blurred image data, the instructions executable by a processor and causing the processor to perform a method comprising:
-
selecting a plurality of first windows from first image data associated with a first image of an object, the first image captured using a first imaging system; selecting a corresponding plurality of second windows from second image data associated with a second image of the object wherein corresponding first and second windows contain a same edge in the object, the second image captured using a second imaging system wherein a comparison of blurring by the first and second imaging systems varies as a function of object depth; for pairs of corresponding first and second windows, estimating the comparison of blurring by the first and second imaging systems based on blurring of the same edge in corresponding first and second windows, wherein estimating the comparison of blurring by the first and second imaging systems comprises; binarizing the same edge in corresponding first and second windows, and determining a blur kernel that approximates blurring of the same binarized edge in corresponding first and second windows, wherein different blur kernels correspond to different object depths; and generating depth information for the object based on said estimated comparisons comprises selecting the object depth that corresponds to the determined blur kernel.
-
Specification