FACE RECOGNITION PERFORMANCE USING ADDITIONAL IMAGE FEATURES

• US 20110081052A1
• Filed: 10/02/2009
• Published: 04/07/2011
• Est. Priority Date: 10/02/2009
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: FotoNation Ireland Limited (Adeia Inc.)
• Inventors: Costache, Gabriel, Drimbarean, Alexandru, Bigioi, Petronel, Corcoran, Peter
• Granted Patent: US 8,379,917 B2
• Time in Patent Office: Days
• US Class Current: 382/118
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: G06V 40/167 using comparisons between temporally consecutive images
• CPC Class Codes: G06V 40/173 face re-identification, e.g. recognising unknown faces across different face tracks

Abstract

A technique is provided for recognizing faces in an image stream using a digital image acquisition device. A first acquired image is received from an image stream. A first face region is detected within the first acquired image having a given size and a respective location within the first acquired image. First faceprint data uniquely identifying the first face region are extracted along with first peripheral region data around the first face region. The first faceprint and peripheral region data are stored, and the first peripheral region data are associated with the first face region. The first face region is tracked until a face lock is lost. A second face region is detected within a second acquired image from the image stream. Second peripheral region data around the second face region are extracted. The second face region is identified upon matching the first and second peripheral region data.

Claims

1. A method of recognizing faces in an image stream using a digital image acquisition device, comprising:

receiving a first acquired image from an image stream;

detecting a first face region within the first acquired image having a given size and a respective location within the first acquired image;

extracting first faceprint data uniquely identifying the first face region along with first peripheral region data around the first face region;

storing the first faceprint and peripheral region data, including associating the first peripheral region data with the first face region;

tracking the first face region until face lock is lost;

detecting a second face region within a second acquired image from the image stream;

extracting second peripheral region data around the second face region; and

identifying the second face region upon matching the first and second peripheral region data.

2. The method of claim 1, wherein the storing of peripheral region data occurs only within volatile memory.

3. The method of claim 1, further comprising providing a database comprising an identifier and associated parameters for each of a number of one or more faces to be recognized;

and selectively applying face recognition using said database to said first face region to provide an identifier for the first face region.

4. The method of claim 1, further comprising extracting second faceprint data uniquely identifying the second face region, and matching the first and second faceprint data, thereby confirming the identifying of the second face region.

5. The method of claim 1, further comprising retrieving and matching texture information from the first and second peripheral region data.

6. The method of claim 1, wherein the detecting and identifying of the second face region occur within two second or less.

7. The method of claim 1, wherein the detecting and identifying of the second face region occur within one second or less.

8. The method of claim 1, further comprising displaying a same identifier with the first and second face regions.

9. The method of claim 8, wherein the identifier comprises a nickname of a person associated with the first and second face regions.

10. The method of claim 1, wherein the image stream comprises two or more relatively low resolution reference images.

11. One or more processor-readable storage media having code embedded therein for programming a processor to perform a method of recognizing faces in an image stream using a digital image acquisition device, wherein the method comprises:

receiving a first acquired image from an image stream;

detecting a first face region within the first acquired image having a given size and a respective location within the first acquired image;

extracting first faceprint data uniquely identifying the first face region along with first peripheral region data around the first face region;

storing the first faceprint and peripheral region data, including associating the first peripheral region data with the first face region;

tracking the first face region until face lock is lost;

detecting a second face region within a second acquired image from the image stream;

extracting second peripheral region data around the second face region; and

identifying the second face region upon matching the first and second peripheral region data.

12. The one or more processor-readable media of claim 11, wherein the storing of peripheral region data occurs only within volatile memory.

13. The one or more processor-readable media of claim 11, wherein the method further comprises providing a database comprising an identifier and associated parameters for each of a number of one or more faces to be recognized;

and selectively applying face recognition using said database to said first face region to provide an identifier for the first face region.

14. The one or more processor-readable media of claim 11, wherein the method further comprises extracting second faceprint data uniquely identifying the second face region, and matching the first and second faceprint data, thereby confirming the identifying of the second face region.

15. The one or more processor-readable media of claim 11, wherein the method further comprises retrieving and matching texture information from the first and second peripheral region data.

16. The one or more processor-readable media of claim 11, wherein the detecting and identifying of the second face region occur within two second or less.

17. The one or more processor-readable media of claim 11, wherein the detecting and identifying of the second face region occur within one second or less.

18. The one or more processor-readable media of claim 11, wherein the method further comprises displaying a same identifier with the first and second face regions.

19. The one or more processor-readable media of claim 18, wherein the identifier comprises a nickname of a person associated with the first and second face regions.

20. The method of claim 11, wherein the image stream comprises two or more relatively low resolution reference images.

21. A portable, digital image acquisition device, comprising a lens and image sensor for acquiring digital images, a processor, and a processor-readable medium having code embedded therein for programming the processor to perform a method of recognizing faces in an image stream using a digital image acquisition device, wherein the method comprises:

receiving a first acquired image from an image stream;

detecting a first face region within the first acquired image having a given size and a respective location within the first acquired image;

extracting first faceprint data uniquely identifying the first face region along with first peripheral region data around the first face region;

storing the first faceprint and peripheral region data, including associating the first peripheral region data with the first face region;

tracking the first face region until face lock is lost;

detecting a second face region within a second acquired image from the image stream;

extracting second peripheral region data around the second face region; and

identifying the second face region upon matching the first and second peripheral region data.

22. The device of claim 21, wherein the storing of peripheral region data occurs only within volatile memory.

23. The device of claim 21, further comprising providing a database comprising an identifier and associated parameters for each of a number of one or more faces to be recognized;

and selectively applying face recognition using said database to said first face region to provide an identifier for the first face region.

24. The device of claim 21, further comprising extracting second faceprint data uniquely identifying the second face region, and matching the first and second faceprint data, thereby confirming the identifying of the second face region.

25. The device of claim 21, further comprising retrieving and matching texture information from the first and second peripheral region data.

26. The device of claim 21, wherein the detecting and identifying of the second face region occur within two second or less.

27. The device of claim 21, wherein the detecting and identifying of the second face region occur within one second or less.

28. The device of claim 21, further comprising displaying a same identifier with the first and second face regions.

29. The device of claim 28, wherein the identifier comprises a nickname of a person associated with the first and second face regions.

30. The device of claim 21, wherein the image stream comprises two or more relatively low resolution reference images.

31. A method of recognizing faces in an image stream using a digital image acquisition device, comprising:

receiving a first acquired image from an image stream;

detecting a first face region within the first acquired image having a given size and a respective location within the first acquired image;

extracting first faceprint data uniquely identifying the first face region along with first peripheral region data around the first face region;

storing the first faceprint and peripheral region data, including associating the first peripheral region data with the first face region;

tracking a first face region combination, including the first face region and peripheral region data, until face lock is lost; and

identifying the first face region.

32. The method of claim 31, further comprising providing a database comprising an identifier and associated parameters for each of a number of one or more faces to be recognized;

and selectively applying face recognition using said database to said first face region combination to provide an identifier for the first face region combination.

33. The method of claim 32, further comprising retrieving and matching texture information from the first peripheral region data for comparing with texture information from a stored peripheral region.

34. The method of claim 31, wherein the image stream comprises two or more relatively low resolution reference images.

35. One or more processor-readable storage media having code embedded therein for programming a processor to perform a method of recognizing faces in an image stream using a digital image acquisition device, wherein the method comprises:

receiving a first acquired image from an image stream;

detecting a first face region within the first acquired image having a given size and a respective location within the first acquired image;

extracting first faceprint data uniquely identifying the first face region along with first peripheral region data around the first face region;

storing the first faceprint and peripheral region data, including associating the first peripheral region data with the first face region;

tracking a first face region combination, including the first face region and peripheral region data, until face lock is lost; and

identifying the first face region.

36. The one or more processor-readable media of claim 35, wherein the method further comprises providing a database comprising an identifier and associated parameters for each of a number of one or more faces to be recognized;

and selectively applying face recognition using said database to said first face region combination to provide an identifier for the first face region combination.

37. The one or more processor-readable media of claim 36, wherein the method further comprises retrieving and matching texture information from the first peripheral region data for comparing with texture information from a stored peripheral region.

38. The one or more processor-readable media of claim 35, wherein the image stream comprises two or more relatively low resolution reference images.

39. A portable, digital image acquisition device, comprising a lens and image sensor for acquiring digital images, a processor, and a processor-readable medium having code embedded therein for programming the processor to perform a method of recognizing faces in an image stream using a digital image acquisition device, wherein the method comprises:

receiving a first acquired image from an image stream;

detecting a first face region within the first acquired image having a given size and a respective location within the first acquired image;

extracting first faceprint data uniquely identifying the first face region along with first peripheral region data around the first face region;

storing the first faceprint and peripheral region data, including associating the first peripheral region data with the first face region;

tracking a first face region combination, including the first face region and peripheral region data, until face lock is lost; and

identifying the first face region.

40. The device of claim 39, wherein the method further comprises providing a database comprising an identifier and associated parameters for each of a number of one or more faces to be recognized;

and selectively applying face recognition using said database to said first face region combination to provide an identifier for the first face region combination.

41. The device of claim 40, wherein the method further comprises retrieving and matching texture information from the first peripheral region data for comparing with texture information from a stored peripheral region.

42. The device of claim 39, wherein the image stream comprises two or more relatively low resolution reference images.