Rule-based segmentation for objects with frontal view in color images

• US 8,682,029 B2
• Filed: 12/12/2008
• Issued: 03/25/2014
• Est. Priority Date: 12/14/2007
• Status: Expired due to Fees

• Current Assignee: Flashfoto, Inc.
• Original Assignee: Flashfoto, Inc.
• Inventors: Piramuthu, Robinson
• Primary Examiner(s): Bayat, Ali
• Application Number:
  US12/735,093
• Publication Number:
  US 20100278426A1
• Time in Patent Office: 1,929 Days
• Field of Search: 382/103, 382/118, 382/164, 382/154, 382/171, 382/173, 382/287
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 40/162 using pixel segmentation or colour matching

Abstract

A method of labeling pixels in an image that represent human skin are detected and one or more regions in the image are identified. Each region may include all or a portion of a human face. Pixels that represent each face in the image are identified using the pixels that represent skin and the regions that include faces of the people, thereby identifying a position of each face in the image. A face mask and a torso map corresponding to each face is determined using determined face positions. The extracted face masks and the torso maps are used to refine a skin map. A person or people map is determined using the skin map and the rough body map.

Claims

1. A method comprising:

forming a face map of a person in an image, wherein the face map is a face region in the image corresponding to a face of the person;

forming a torso map of the person in the image based on the face map, wherein the torso map is a torso region in the image corresponding to a torso of the person; and

forming a person map based on the face map and the torso map, wherein the person map is a person region in the image corresponding to the person, wherein the person map forms a segmented image of the person in the image, and wherein forming the person map includes;

combining the torso map and the face map to form an initial person map;

determining whether an additional region in the image includes a color matching a color in the torso map or in the face map; and

combining the additional region with the initial person map in response to finding a color match.

2. The method of claim 1, wherein forming the face map comprises:

determining an initial face map for the face of the person, wherein the initial face map is a region in the image that corresponds to a position of the face; and

determining whether to combine an additional region in the image with the initial face map.

3. The method of claim 1, wherein the face map includes skin pixels.

4. The method of claim 1, wherein determining the torso map comprises:

determining an initial torso map for the torso of the person, wherein the initial torso map is positioned below the face map; and

determining whether to combine an additional region in the image with the initial torso map.

5. The method of claim 4, wherein determining whether to combine the additional region with the initial torso map comprises:

determining whether the additional region and the initial torso map include similar colors.

6. The method of claim 4, wherein determining whether to combine the additional region with the initial torso map comprises:

determining whether the additional region is adjacent to the initial torso map.

7. The method of claim 1 further comprising:

determining whether to remove a subset of the torso region within the torso map.

8. The method of claim 7, wherein determining whether to remove the subset of the torso region comprises:

determining whether a position of the subset of the torso region is at a top edge of the image.

9. The method of claim 7, wherein determining whether to remove the subset of the torso region comprises:

determining whether a distance between the subset of the torso region and the face map is greater than a certain threshold.

10. The method of claim 7, wherein determining whether to remove the subset of the torso region comprises:

determining whether a size of the subset of the torso region with respect to the face map is greater than a certain size.

11. A method comprising:

forming a face map of a person in an image, wherein the face map is a face region in the image corresponding to a face of the person;

forming a torso map of the person in the image based on the face map, wherein the torso map is a torso region in the image corresponding to a torso of the person;

forming a person map based on the face map and the torso map, wherein the person map is a person region in the image corresponding to the person, wherein the person map forms a segmented image of the person in the image; and

determining whether to remove a subset of the torso region within the torso map, wherein determining whether to remove the subset of the torso region comprises determining whether a distance between the subset of the torso region and the face map is greater than a certain threshold.

12. The method of claim 11, wherein forming the face map comprises:

determining an initial face map for the face of the person, wherein the initial face map is a region in the image that corresponds to a position of the face; and

determining whether to combine an additional region in the image with the initial face map.

13. The method of claim 11, wherein determining the torso map comprises:

determining an initial torso map for the torso of the person, wherein the initial torso map is positioned below the face map; and

determining whether to combine an additional region in the image with the initial torso map.

14. The method of claim 13, wherein determining whether to combine the additional region with the initial torso map comprises:

determining whether the additional region and the initial torso map include similar colors.

15. The method of claim 13, wherein determining whether to combine the additional region with the initial torso map comprises:

determining whether the additional region is adjacent to the initial torso map.

16. A method comprising:

forming a face map of a person in an image, wherein the face map is a face region in the image corresponding to a face of the person;

forming a torso map of the person in the image based on the face map, wherein the torso map is a torso region in the image corresponding to a torso of the person;

forming a person map based on the face map and the torso map, wherein the person map is a person region in the image corresponding to the person, wherein the person map forms a segmented image of the person in the image; and

determining whether to remove a subset of the torso region within the torso map, wherein determining whether to remove the subset of the torso region comprises determining whether a size of the subset of the torso region with respect to the face map is greater than a certain size.

17. The method of claim 16, wherein forming the face map comprises:

determining an initial face map for the face of the person, wherein the initial face map is a region in the image that corresponds to a position of the face; and

determining whether to combine an additional region in the image with the initial face map.

18. The method of claim 16, wherein determining the torso map comprises:

determining an initial torso map for the torso of the person, wherein the initial torso map is positioned below the face map; and

determining whether to combine an additional region in the image with the initial torso map.

19. The method of claim 18, wherein determining whether to combine the additional region with the initial torso map comprises:

determining whether the additional region and the initial torso map include similar colors.

20. The method of claim 18, wherein determining whether to combine the additional region with the initial torso map comprises:

determining whether the additional region is adjacent to the initial torso map.