Laying out multiple images

• US 8,509,566 B2
• Filed: 06/01/2012
• Issued: 08/13/2013
• Est. Priority Date: 04/05/2007
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Jin, Hailin, Chien, Jen-Chan
• Primary Examiner(s): Wang, Claire X
• Assistant Examiner(s): Thirugnanam, Gandhi
• Application Number:
  US13/486,882
• Publication Number:
  US 20120243803A1
• Time in Patent Office: 438 Days
• Field of Search: 382/284
• US Class Current: 382/284
• CPC Class Codes: G06T 3/4038 Image mosaicing, e.g. composing plane images from plane sub-images
• CPC Class Codes: G06T 7/30 Determination of transform parameters for the alignment of images, i.e. image registration

Abstract

Systems, methods, and apparatuses, including computer program products, are provided for re-layout of composite images. In some implementations, a method includes identifying geometric transformations corresponding to multiple images from a collection of images, where a geometric transformation reorients a corresponding image in relation to a common reference frame when applied and identifying a reference image for the multiple images in the collection of images. The method also includes determining overlapping image regions for the multiple images starting from the reference image, the determining based on the identified geometric transformations, determining additional transformations of a specified type for the multiple images based on the overlapping image regions, where an additional transformation lays out a corresponding image in relation to the reference image when applied, and making the additional transformations available for further processing and output with respect to the collection of images.

Claims

1. A method comprising:

identifying geometric transformations corresponding to multiple images from a collection of images, where a geometric transformation reorients a corresponding image in relation to a common reference frame when applied;

identifying a reference image for the multiple images in the collection of images;

determining overlapping image regions for the multiple images starting from the reference image and proceeding according to an order based on a distance measure between images of the multiple images, the distance measure between a particular pair of images of the multiple images being based on a measure of distortion between the pair of images;

determining additional transformations of a specified type for the multiple images based on the overlapping image regions; and

laying out the multiple images.

2. The method of claim 1, wherein determining overlapping image regions for the multiple images further comprises:

identifying a relative transformation for a second image relative to a first image of the multiple images using the identified geometric transformations corresponding to the first image and the second image; and

identifying the overlapping image region of the first image and the second image when the relative transformation is applied to the second image.

3. The method of claim 2, wherein identifying the relative transformation include determining a projective transformation for the second image relative to the first image that projects the second image to a reference frame of the first image.

4. The method of claim 3, wherein for second image i and the first image j, the projective transformation dH_ji is provided by:

dH_ji=(H_j)<sup>−

1</sup>H_i, where (H_j)<sup>−

1</sup> is an inverse of the first projective transformation from the reference frame of image j to a common reference frame and H_i is the first projective transformation from the reference frame of image i to the common reference frame.

5. The method of claim 2, wherein identifying the overlapping image region includes identifying an overlapping polygon for the projected second image to a reference frame of the first image.

6. The method of claim 2, wherein the first image is the reference image and the second image is an image of the multiple images having a shortest distance to the reference image.

7. A computer program product, encoded on a non-transitory computer-readable medium, operable to cause data processing apparatus to perform operations comprising:

identifying geometric transformations corresponding to multiple images from a collection of images, where a geometric transformation reorients a corresponding image in relation to a common reference frame when applied;

identifying a reference image for the multiple images in the collection of images;

determining overlapping image regions for the multiple images starting from the reference image and proceeding according to an order based on a distance measure between images of the multiple images, the distance measure between a particular pair of images of the multiple images being based on a measure of distortion between the pair of images;

determining additional transformations of a specified type for the multiple images based on the overlapping image regions; and

laying out the multiple images.

8. The computer program product of claim 7, wherein the relationships between images of the multiple images are represented using a connectivity graph, wherein images are represented as individual nodes an edges linking nodes together represent overlapping images.

9. The computer program product of claim 8, wherein the length of an edge between two nodes, representing a distance between two overlapping images, varies based on a measure of distortion between the pair of images.

10. The computer program product of claim 9, wherein the distance between a pair of overlapping images (i,j) is computed according to:

Distance_ij=log(Distance(i,j)),

where Distance(i, j) is a sum of the distortion for image j when image i is set as the reference image and the distortion for image i when image j is set as the reference image.

11. The computer program product of claim 8, wherein identifying the reference image includes computing a maximal distance from each image to all other images of the multiple images and determining at least one image having a minimum maximal distance.

12. The computer program product of claim 8, wherein identifying the reference image includes determining an image with a minimum average distance to all other images of the multiple images.

13. A system comprising:

one or more computers configured to perform operations comprising;

identifying geometric transformations corresponding to multiple images from a collection of images, where a geometric transformation reorients a corresponding image in relation to a common reference frame when applied;

identifying a reference image for the multiple images in the collection of images;

determining overlapping image regions for the multiple images starting from the reference image and proceeding according to an order based on a distance measure between images of the multiple images, the distance measure between a particular pair of images of the multiple images being based on a measure of distortion between the pair of images;

determining additional transformations of a specified type for the multiple images based on the overlapping image regions; and

laying out the multiple images.

14. The system of claim 13, wherein the additional transformations are two-dimensional (2D) translations and determining the 2D translation for the second image includes determining an incremental 2D translation relative to a translation of the first image.

15. The system of claim 13, wherein the additional transformations are 2D translations and determining the 2D translations for the multiple images based on the overlapping image regions includes identifying center points for each of the overlapping image regions.

16. The system of claim 15, wherein identifying the center point of an overlapping image region comprises:

identifying coordinates for a plurality of vertices of the overlapping region; and

calculating a mean of the coordinates for the plurality of vertices.

17. The system of claim 13, wherein the additional transformations are 2D translation and rotation transformations, and where determining the translation and rotation for one or more of the multiple images includes using two or more points in the determined overlapping regions.

18. The system of claim 13, wherein the additional transformations are 2D translation, rotation and scaling transformations and where determining the translation, rotation and scaling for one or more images of the multiple images includes using two or more points in the determined overlapping regions.

19. The system of claim 13, further configured to perform operations comprising performing user specified types of additional transformations where the type of additional transformations is user specified.

20. The system of claim 13, further configured to perform operations comprising:

displaying a composite layout for the multiple images.