Active segmentation for groups of images

• US 8,737,739 B2
• Filed: 10/19/2011
• Issued: 05/27/2014
• Est. Priority Date: 06/11/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Tang, Xiaoou, Yang, Qiong
• Primary Examiner(s): Bayat, Ali
• Application Number:
  US13/276,881
• Publication Number:
  US 20120093411A1
• Time in Patent Office: 951 Days
• Field of Search: 382/173, 382/167, 382/171
• US Class Current: 382/173
• CPC Class Codes: G06V 10/248 by interactive preprocessing or interactive shape modelling, e.g. feature points assigned by a user
• CPC Class Codes: G06V 10/267 by performing operations on regions, e.g. growing, shrinking or watersheds
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns

Abstract

Systems and methods of segmenting images are disclosed. The similarity of images in a set of images is compared. A group of images is selected from the set of images. The images in the group of images are selected based on compared similarities among the images. An informative image is selected from the group of images. User-defined semantic information of the informative image is received. The group of images is modeled as a graph. Each image in the group of images denotes a node in the graph. Edges of the graph denote a foreground or background relationship between images. One or more images in the group of images may be automatically segmented by propagating semantic information of the informative image to images in the group having a graph node corresponding to the informative image. Segmentation results can be refined according to user provided image semantics.

Claims

1. A computer-implemented method comprising:

by one or more processors, selecting an informative image that is representative of a group of similar images;

automatically modeling the group of similar images to construct a graph including one or more edges that indicate at least one of a foreground relationship or a background relationship between two or more images in the group of similar images; and

segmenting one or more images in the group of similar images by propagating semantic information of the informative image to at least one image in the group of similar images, the propagating based on a relationship in the graph between the informative image and the at least one image.

2. The method of claim 1, further comprising receiving the semantic information for the informative image.

3. The method of claim 1, wherein the semantic information is user-specified for the informative image.

4. The method of claim 1, the graph including a node for each image in the group of similar images.

5. The method of claim 1, further comprising selecting the group of similar images from a plurality of images, based on one or more similarities of a word distribution for each of the plurality of images.

6. The method of claim 1, wherein the selecting the informative image includes selecting an image in the group of similar images with a higher number of non-overlapping words in relation to other images in the group of similar images.

7. The method of claim 1, wherein the relationship in the graph between the informative image and the at least one image is indicated by an edge of the one or more edges between the informative image and the at least one image.

8. A system comprising:

one or more processing units; and

at least one software module executable by the one or more processing units to;

select an informative image from a group of similar images, the informative image being representative of the group of similar images;

construct a graph by modeling the group of similar images, the graph including one or more edges that indicate at least one of a foreground relationship or a background relationship between images in the group of similar images; and

segment one or more images in the group of similar images by propagating semantic information of the informative image to at least one image in the group of similar images, the propagating based at least in part on a relationship in the graph between the informative image and the at least one image.

9. The system of claim 8, wherein the at least one software module further operates to select the group of similar images from a plurality of images, based on one or more similarities of a word histogram for each of the plurality of images.

10. The system of claim 8, wherein the at least one software module further operates to select the group of similar images from a plurality of images, based on at least one of a foreground similarity or a background similarity.

11. The system of claim 8, wherein the at least one software module further operates to receive the semantic information for the informative image.

12. The system of claim 8, the graph including a node for each image in the group of similar images.

13. The system of claim 8, wherein the selection of the informative image includes selecting an image in the group of similar images with a higher number of non-overlapping words in relation to other images in the group of similar images.

14. The system of claim 8, wherein the relationship in the graph between the informative image and the at least one image is indicated by an edge of the one or more edges between the informative image and the at least one image.

15. One or more computer-readable storage memory devices storing computer-executable instructions that instruct a processor to perform actions comprising:

selecting an informative image that is typical of a group of images;

modeling the group of images to construct a graph including one or more edges that indicate at least one of a foreground relationship or a background relationship between images in the group of images; and

segmenting one or more images in the group of images by propagating semantic information of the informative image to at least one image in the group of images.

16. The one or more computer-readable storage memory devices of claim 15, the propagating based at least in part on a relationship in the graph between the informative image and the at least one image.

17. The one or more computer-readable storage memory devices of claim 16, wherein the relationship in the graph between the informative image and the at least one image is indicated by an edge of the one or more edges between the informative image and the at least one image.

18. The one or more computer-readable storage memory devices of claim 15, the actions further comprising selecting the group of images from a plurality of images, based on one or more similarities of a word distribution for each of the plurality of images.

19. The one or more computer-readable storage memory devices of claim 15, the graph including a node for each image in the group of images.

20. The one or more computer-readable storage memory devices of claim 15, wherein the selecting the informative image includes selecting an image in the group of images with a higher number of non-overlapping words in relation to other images in the group of images.