Automatic face and skin beautification using face detection

• US 9,007,480 B2
• Filed: 07/30/2009
• Issued: 04/14/2015
• Est. Priority Date: 07/30/2008
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: FotoNation Limited (Adeia Inc.)
• Inventors: Ciuc, Mihai, Capata, Adrian, Mocanu, Valentin, Pososin, Alexei, Florea, Corneliu, Corcoran, Peter
• Primary Examiner(s): Tran, Sinh
• Assistant Examiner(s): Nazrul, Shahbaz
• Application Number:
  US12/512,843
• Publication Number:
  US 20100026833A1
• Time in Patent Office: 2,084 Days
• Field of Search: 348/222.1, 382/118, 382/254
• US Class Current: 348/222.1
• CPC Class Codes: G06T 11/40 Filling a planar surface by adding surface attributes, e.g. colour or texture
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06V 40/16 Human faces, e.g. facial parts, sketches or expressions
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: G06V 40/171 Local features and components; Facial parts eye characteristics G06V40/18; Occluding parts, e.g. glasses; Geometrical relationships
• CPC Class Codes: H04N 2101/00 Still video cameras
• CPC Class Codes: H04N 23/10 for generating image signals from different wavelengths
• CPC Class Codes: H04N 23/40 Circuit details for pick-up tubes
• CPC Class Codes: H04N 23/80 Camera processing pipelines; Components thereof
• CPC Class Codes: H04N 23/95 Computational photography systems, e.g. light-field imaging systems
• CPC Class Codes: H04N 5/2621 Cameras specially adapted for the electronic generation of special effects during image pickup, e.g. digital cameras, camcorders, video cameras having integrated special effects capability

Abstract

Sub-regions within a face image are identified to be enhanced by applying a localized smoothing kernel to luminance data corresponding to the sub-regions of the face image. An enhanced face image is generated including an enhanced version of the face that includes certain original pixels in combination with pixels corresponding to the one or more enhanced sub-regions of the face.

Claims

1. A method of enhancing an appearance of a face within a digital image, comprising one or more processors performing the steps of:

acquiring a digital image of a scene including a face,

wherein acquiring the digital image includes

capturing the image using a lens and an image sensor, or

receiving said image following capture by a device that includes a lens and an image sensor, or

a combination thereof;

identifying the face within the digital image;

identifying within the face two or more skin tone portions to be enhanced with localized luminance smoothing;

applying one or more localized luminance smoothing kernels to at least one skin tone portion of the one or more skin tone portions identified within the face to produce one or more enhanced skin tone portions of the face,

wherein applying the one or more localized luminance smoothing kernels comprises;

determining an amount of difference between a luminance value of a current pixel of the at least one skin tone portion and a local average luminance value of the at least one skin tone portion;

determining whether the amount of difference between the luminance value of the current pixel and the local average luminance is less than a particular threshold amount;

determining a weight of smoothing to apply to the current pixel by the one or more localized smoothing kernels by;

increasing a prior smoothing weight if the amount of difference between the luminance value of the current pixel and the local average luminance is less than the particular threshold amount;

or

decreasing a prior smoothing weight if the amount of difference between the luminance value of the current pixel and the local average luminance is not less than the particular threshold amount;

applying the one or more localized luminance smoothing kernels only to luminance data of the at least one skin tone portion;

generating an enhanced image including an enhanced version of the face comprising certain original or otherwise processed pixels in combination with pixels corresponding to the one or more enhanced skin tone portions of the face; and

displaying, transmitting, communicating or digitally storing or otherwise outputting the enhanced image or a further processed version, or combinations thereof.

2. The method of claim 1, wherein the localized luminance smoothing comprises blurring or averaging luminance data, or a combination thereof.

3. The method of claim 1, further comprising:

applying one or more localized color smoothing kernels to one or more blemish regions identified within one or more of the skin tone portions or one or more non-skin tone face features including one or both eyes or a mouth, or combinations thereof, and

wherein the enhanced version of the face of the enhanced image further comprises pixels modified from original pixels of the face at least by localized color smoothing.

4. The method of claim 1, further comprising:

applying noise reduction or enhancement, or both, to the one or more skin tone portions or one or more non-skin tone face features including one or both eyes or a mouth, or combinations thereof, and

wherein the enhanced version of the face of the enhanced image further comprises pixels modified from original pixels of the face at least by localized noise reduction or enhancement, or both.

5. The method of claim 1, further comprising:

determining certain non-skin tone pixels within the one or more skin tone portions that do not comprise a threshold skin tone, and

removing, replacing, reducing an intensity of, or modifying a color of said certain non-skin tone pixels, or combinations thereof.

6. The method of claim 1, wherein:

enhanced pixels of the one or more enhanced skin tone portions comprise enhanced intensities which comprise one or more functions of a relationship between original pixel intensities and local average intensities within the one or more original or enhanced skin tone portions, or combinations thereof.

7. The method of claim 1, further comprising:

detecting one or more mouth or eye regions, or combinations thereof, within the face, and

identifying and enhancing a natural color of one or more sub-regions within the one or more mouth or eye regions, including one or more teeth, lips, tongues, eye whites, eye brows, iris'"'"'s, eye lashes, or pupils, or combinations thereof.

8. The method of claim 1, further comprising:

classifying the face according to its age based on comparing one or more default image attribute values with one or more determined values; and

adjusting one or more camera acquisition or post-processing parameters, or combinations thereof, based on the classifying of the face according to its age.

24. The one or more processor-readable media of claim 1, wherein the method further comprises classifying the face according to its age based on comparing one or more default image attribute values with one or more determined values;

and adjusting one or more camera acquisition or post-processing parameters, or combinations thereof, based on the classifying of the face according to its age.

9. A digital image acquisition device, comprising a lens, an image sensor and a processor, and a processor-readable memory having embodied therein processor-readable code for programming the processor to perform a method of enhancing an appearance of a face within a digital image, wherein the method comprises:

acquiring a digital image of a scene including a face,

wherein acquiring the digital image includes

capturing the image using a lens and an image sensor, or

receiving said image following capture by a device that includes a lens and an image sensor, or

a combination thereof;

identifying the face within the digital image;

identifying within the face two or more skin tone portions to be enhanced with localized luminance smoothing;

applying one or more localized luminance smoothing kernels to at least one skin tone portion of the one or more skin tone portions identified within the face to produce one or more enhanced skin tone portions of the face,

wherein applying the one or more localized luminance smoothing kernels comprises;

determining an amount of difference between a luminance value of a current pixel of the at least one skin tone portion and a local average luminance value of the at least one skin tone portion;

determining whether the amount of difference between the luminance value of the current pixel and the local average luminance is less than a particular threshold amount;

determining a weight of smoothing to apply to the current pixel by the one or more localized smoothing kernels by;

increasing a prior smoothing weight if the amount of difference between the luminance value of the current pixel and the local average luminance is less than the particular threshold amount;

or

decreasing a prior smoothing weight if the amount of difference between the luminance value of the current pixel and the local average luminance is not less than the particular threshold amount;

applying the one or more localized luminance smoothing kernels only to luminance data of the at least one skin tone portion;

generating an enhanced image including an enhanced version of the face comprising certain original or otherwise processed pixels in combination with pixels corresponding to the one or more enhanced skin tone portions of the face; and

displaying, transmitting, communicating or digitally storing or otherwise outputting the enhanced image or a further processed version, or combinations thereof.

10. The device of claim 9, wherein the localized luminance smoothing comprises blurring or averaging luminance data, or a combination thereof.

11. The device of claim 9, wherein the method further comprises applying one or more localized color smoothing kernels to one or more blemish regions identified within one or more of the skin tone portions or one or more non-skin tone face features including one or both eyes or a mouth, or combinations thereof, and wherein the enhanced version of the face of the enhanced image further comprises pixels modified from original pixels of the face at least by localized color smoothing.

12. The device of claim 9, wherein the method further comprises applying noise reduction or enhancement, or both, to the one or more skin tone portions or one or more non-skin tone face features including one or both eyes or a mouth, or combinations thereof, and wherein the enhanced version of the face of the enhanced image further comprises pixels modified from original pixels of the face at least by localized noise reduction or enhancement, or both.

13. The device of claim 9, wherein the method further comprises determining certain non-skin tone pixels within the one or more skin tone portions that do not comprise a threshold skin tone, and removing, replacing, reducing an intensity of, or modifying a color of said certain non-skin tone pixels, or combinations thereof.

14. The device of claim 9, wherein enhanced pixels of the one or more enhanced skin tone portions comprise enhanced intensities which comprise one or more functions of a relationship between original pixel intensities and local average intensities within the one or more original or enhanced skin tone portions, or combinations thereof.

15. The device of claim 9, wherein the method further comprises detecting one or more mouth or eye regions, or combinations thereof, within the face, and identifying and enhancing a natural color of one or more sub-regions within the one or more mouth or eye regions, including one or more teeth, lips, tongues, eye whites, eye brows, iris'"'"'s, eye lashes, or pupils, or combinations thereof.

16. The device of claim 9, wherein the method further comprises classifying the face according to its age based on comparing one or more default image attribute values with one or more determined values;

and adjusting one or more camera acquisition or post-processing parameters, or combinations thereof, based on the classifying of the face according to its age.

17. One or more non-transitory processor-readable media having embodied therein code for programming one or more processors to perform a method of enhancing an appearance of a face within a digital image, wherein the method comprises:

acquiring a digital image of a scene including a face,

wherein acquiring the digital image includes

capturing the image using a lens and an image sensor, or

receiving said image following capture by a device that includes a lens and an image sensor, or

a combination thereof;

identifying the face within the digital image;

identifying within the face two or more skin tone portions to be enhanced with localized luminance smoothing;

applying one or more localized luminance smoothing kernels to at least one skin tone portion of the one or more skin tone portions identified within the face to produce one or more enhanced skin tone portions of the face,

wherein applying the one or more localized luminance smoothing kernels comprises;

determining an amount of difference between a luminance value of a current pixel of the at least one skin tone portion and a local average luminance value of the at least one skin tone portion;

determining whether the amount of difference between the luminance value of the current pixel and the local average luminance is less than a particular threshold amount;

determining a weight of smoothing to apply to the current pixel by the one or more localized smoothing kernels by;

increasing a prior smoothing weight if the amount of difference between the luminance value of the current pixel and the local average luminance is less than the particular threshold amount;

or

decreasing a prior smoothing weight if the amount of difference between the luminance value of the current pixel and the local average luminance is not less than the particular threshold amount;

applying the one or more localized luminance smoothing kernels only to luminance data of the at least one skin tone portion;

generating an enhanced image including an enhanced version of the face comprising certain original or otherwise processed pixels in combination with pixels corresponding to the one or more enhanced skin tone portions of the face; and

displaying, transmitting, communicating or digitally storing or otherwise outputting the enhanced image or a further processed version, or combinations thereof.

18. The one or more processor-readable media of claim 17, wherein the localized luminance smoothing comprises blurring or averaging luminance data, or a combination thereof.

19. The one or more processor-readable media of claim 17, wherein the method further comprises applying one or more localized color smoothing kernels to one or more blemish regions identified within one or more of the skin tone portions or one or more non-skin tone face features including one or both eyes or a mouth, or combinations thereof, and wherein the enhanced version of the face of the enhanced image further comprises pixels modified from original pixels of the face at least by localized color smoothing.

20. The one or more processor-readable media of claim 17, wherein the method further comprises applying noise reduction or enhancement, or both, to the one or more skin tone portions or one or more non-skin tone face features including one or both eyes or a mouth, or combinations thereof, and wherein the enhanced version of the face of the enhanced image further comprises pixels modified from original pixels of the face at least by localized noise reduction or enhancement, or both.

21. The one or more processor-readable media of claim 17, wherein the method further comprises determining certain non-skin tone pixels within the one or more skin tone portions that do not comprise a threshold skin tone, and removing, replacing, reducing an intensity of, or modifying a color of said certain nonskin tone pixels, or combinations thereof.

22. The one or more processor-readable media of claim 17, wherein enhanced pixels of the one or more enhanced skin tone portions comprise enhanced intensities which comprise one or more functions of a relationship between original pixel intensities and local average intensities within the one or more original or enhanced skin tone portions, or combinations thereof.

23. The one or more processor-readable media of claim 17, wherein the method further comprises detecting one or more mouth or eye regions, or combinations thereof, within the face, and identifying and enhancing a natural color of one or more sub-regions within the one or more mouth or eye regions, including one or more teeth, lips, tongues, eye whites, eye brows, iris'"'"'s, eye lashes, or pupils, or combinations thereof.