Camera calibration and automatic adjustment of images
Techniques and apparatus for automatic upright adjustment of digital images. An automatic upright adjustment technique is described that may provide an automated approach for straightening up slanted features in an input image to improve its perceptual quality. This correction may be referred to as upright adjustment. A set of criteria based on human perception may be used in the upright adjustment. A reprojection technique that implements an optimization framework is described that yields an optimal homography for adjustment based on the criteria and adjusts the image according to new camera parameters generated by the optimization. An optimization-based camera calibration technique is described that simultaneously estimates vanishing lines and points as well as camera parameters for an image; the calibration technique may, for example, be used to generate estimates of camera parameters and vanishing points and lines that are input to the reprojection technique.
1. In a digital medium environment to adjust alignment of at least one feature in a digital image of a scene consistent with human perception, a computing device comprising:
a memory configured to maintain the digital image;
a processor system configured to execute an image processing application that is implemented to perform operations including:
extracting line segments from the digital image;
identifying an orthogonal frame of the digital image based on the extracted line segments, the orthogonal frame having an additional horizontal direction that shares a same vertical direction;
estimating an eye-level of the scene based on the orthogonal frame having an additional horizontal direction; and
calibrating camera parameters as a basis to adjust alignment of the at least one feature in relation to an axis of the digital image according to one or more human perception-based criteria that include the estimated eye-level of the scene.
2. A computing device as described in claim 1 , wherein the orthogonal frame follows an Atlanta world assumption in which the image includes multiple orthogonal directions that do not align, one to another.
3. A computing device as described in claim 1 , wherein the axis is a vertical or horizontal axis and the at least one feature is substantially aligned with the vertical or horizontal axis subsequent to calibrating the camera parameters.
4. A computing device as described in claim 1 , wherein the extracting includes extracting Manhattan directions and additional horizontal vanishing points.
5. A computing device as described in claim 1 , wherein the identifying further includes determining vanishing lines for each vanishing point.
6. A computing device as described in claim 1 , wherein said calibrating camera parameters includes setting an x-axis focal length and a y-axis focal length, the y-axis focal length being different than the x-axis focal length.
7. A computing device as described in claim 1 , wherein the camera parameters include an intrinsic camera parameter matrix and three dimensional rotation matrices.
8. A computing device as described in claim 1 , wherein the image processing application is further implemented to generate an adjusted digital image based on calibrating the camera parameters.
9. In a digital medium environment to adjust alignment of at least one feature in a digital image of a scene consistent with human perception, a method implemented by a computing device, the method comprising:
extracting, by the computing device, line segments from the digital image;
identifying, by the computing device, an orthogonal frame of the digital image based on the extracted line segments, the orthogonal frame having an additional horizontal direction that shares a same vertical direction;
estimating, by the computing device, an eye-level of the scene based on the orthogonal frame having an additional horizontal direction; and
calibrating, by the computing device, camera parameters as a basis to adjust alignment of the at least one feature in relation to an axis of the digital image according to one or more human perception-based criteria that include the estimated eye-level of the scene.
10. A method as described in claim 9 , wherein the orthogonal frame follows an Atlanta world assumption in which the image includes multiple orthogonal directions that do not align, one to another.
11. A method as described in claim 9 , wherein the axis is a vertical or horizontal axis and the at least one feature is substantially aligned with the vertical or horizontal axis subsequent to calibrating the camera parameters.
12. A method as described in claim 9 , wherein the extracting includes extracting Manhattan directions and additional horizontal vanishing points.
13. A method as described in claim 9 , the identifying further comprising determining vanishing lines for each vanishing point.
14. A method as described in claim 9 , further comprising generating an adjusted digital image based on calibrating the camera parameters.
15. In a digital medium environment to adjust alignment of at least one feature in a digital image of a scene consistent with human perception, a system comprising:
means for identifying an orthogonal frame of the digital image based on line segments extracted from the digital image, the orthogonal frame having an additional horizontal direction that shares a same vertical direction;
means for estimating an eye-level of the scene based on the orthogonal frame having an additional horizontal direction; and
means for calibrating camera parameters as a basis to adjust alignment of the at least one feature in relation to an axis of the digital image according to one or more human perception-based criteria that include the estimated eye-level of the scene.
16. A system as described in claim 15 , wherein the orthogonal frame follows an Atlanta world assumption in which the image includes multiple orthogonal directions that do not align, one to another.
17. A system as described in claim 15 , wherein the axis is a vertical or horizontal axis and the at least one feature is substantially aligned with the vertical or horizontal axis subsequent to calibrating the camera parameters.
18. A system as described in claim 15 , the means for identifying an orthogonal frame further comprising means for identifying Manhattan directions and additional horizontal vanishing points.
19. A system as described in claim 15 , the means for identifying an orthogonal frame further comprising means for determining vanishing lines for each vanishing point.
20. A system as described in claim 15 , further comprising means for generating an adjusted digital image based on calibrating the camera parameters.