Face tracking in a camera processor

• US 8,155,397 B2
• Filed: 09/26/2007
• Issued: 04/10/2012
• Est. Priority Date: 09/26/2007
• Status: Expired due to Fees

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: DigitalOptics Corporation Europe Limited (Adeia Inc.)
• Inventors: Bigioi, Petronel, Gangea, Mihnea, Petrescu, Stefan, Corcoran, Peter, Pososin, Alexei
• Primary Examiner(s): Tucker, Wesley
• Assistant Examiner(s): WOLDEMARIAM, AKLILU K
• Application Number:
  US11/861,854
• Publication Number:
  US 20090080713A1
• Time in Patent Office: 1,658 Days
• Field of Search: 382/118, 382/190, 382/282, 382/209, 382/167, 382/274, 348/169, 348/222.1, 348/241
• US Class Current: 382/118
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

A method operable in a digital image acquisition system having no photographic film is provided. The method comprises receiving a relatively low resolution image of a scene from an image stream, wherein the scene potentially includes one or more faces. At least one high quality face classifier is applied to the image to identify relatively large and medium sized face regions and at least one relaxed face classifier is applied to the image to identify relatively small sized face regions. A relatively high resolution image of nominally the same scene is received and at least one high quality face classifier is applied to the identified small sized face regions in the higher resolution version of said image.

Claims

1. A method operable in a digital image acquisition system having no photographic film, comprising:

using a processor in performing the following;

a) receiving a relatively low resolution image of a scene from an image stream, said scene including one or more faces;

b) applying at least one relatively long face classifier to said image to identify any relatively large sized face regions;

c) applying at least one relatively short face classifier to said image to identify one or more relatively small sized face regions;

d) receiving a relatively high resolution image of approximately the same scene; and

e) applying at least one relatively long face classifier to at least one of said one or more identified small sized face regions in said relatively high resolution image.

2. The method according to claim 1, further comprising performing steps a) to c) on a first processor and performing steps d) and e) on a separate second processor.

3. The method according to claim 2, further comprising performing value added applications on said high resolution image on said separate second processor.

4. The method of claim 1 wherein each of steps b) and c) include providing information including face size, face location, and an indication of a probability of said image including a face at or in the vicinity of said face region.

5. The method of claim 1, wherein step b) or step c), or both, include providing information including face size, face location, or an indication of a probability of said image including a face at or in the vicinity of said face region, or combinations thereof.

6. The method of claim 5 further comprising generating a weighting based on said information.

7. The method according to claim 5 comprising adjusting one or more image acquisition parameters of a subsequent image in said image stream based on said information.

8. The method according to claim 7 wherein said one or more image acquisition parameters that are adjusted include focus, exposure or white balance, or combinations thereof.

9. The method according to claim 7 wherein said subsequent image in said stream is a preview image.

10. The method according to claim 7 wherein said subsequent image in said stream is a main acquired image.

11. The method according to claim 7 further comprising displaying said subsequent image to a user.

12. The method according to claim 1, wherein said at least one relatively long face classifier comprises a classifier with a relatively high threshold for accepting a face, or said relatively short classifier comprises a classifier with a relatively low threshold for accepting a face, or both.

13. A digital image acquisition apparatus, comprising a first processor coupled to an imaging sensor, and a second processor coupled to said first processor, wherein said first processor is arranged to provide an acquired image to said second processor and said second processor is arranged to store said image, and

wherein said first processor is arranged to apply at least one relatively long face classifier to a relatively low resolution image of a scene from an image stream, said scene including one or more faces, to identify any relatively large sized face regions, and to apply at least one relatively short face classifier to said image to identify one or more relatively small sized face regions;

and

wherein said second processor is arranged to receive a relatively high resolution image of approximately the same scene and to apply at least one relatively long face classifier to at least one of said one or more identified small sized face regions in said relatively high resolution image.

14. One or more non-transitory processor-readable storage devices having program code embodied therein for programming one or more processors to perform a method operable in a digital image acquisition system having no photographic film, wherein the method comprises:

a) receiving a relatively low resolution image of a scene from an image stream, said scene including one or more faces;

b) applying at least one relatively long face classifier to said image to identify any relatively large sized face regions;

c) applying at least one relatively short face classifier to said image to identify one or more relatively small sized face regions;

d) receiving a relatively high resolution image of approximately the same scene; and

e) applying at least one relatively long face classifier to at least one of said one or more identified small sized face regions in said relatively high resolution image.

15. The one or more storage devices of claim 14, wherein the method further comprises performing steps a) to c) on a first processor and performing steps d) and e) on a separate second processor.

16. The one or more storage devices of claim 15, wherein the method further comprises performing value added applications on said high resolution image on said separate second processor.

17. The one or more storage devices of claim 14, wherein each of steps b) and c) include providing information including face size, face location, and an indication of a probability of said image including a face at or in the vicinity of said face region.

18. The one or more storage devices of claim 14, wherein step b) or step c), or both, include providing information including face size, face location, or an indication of a probability of said image including a face at or in the vicinity of said face region, or combinations thereof.

19. The one or more storage devices of claim 18, wherein the method further comprises generating a weighting based on said information.

20. The one or more storage devices of claim 18, wherein the method further comprises adjusting image acquisition parameters of a subsequent image in said image stream based on said information.

21. The one or more storage devices of claim 20, wherein said adjusted image acquisition parameters include focus, exposure or white balance, or combinations thereof.

22. The one or more storage devices according to claim 20, wherein said subsequent image in said stream comprises a preview image.

23. The one or more storage devices of claim 20, wherein said subsequent image in said stream comprises a main acquired image.

24. The one or more storage devices of claim 20, wherein the method further comprises displaying said subsequent image to a user.

25. The one or more storage devices of claim 14, wherein said at least one relatively long face classifier comprises a classifier with a relatively high threshold for accepting a face, or said relatively short classifier comprises a classifier with a relatively low threshold for accepting a face, or both.