FACE COMPONENT EXTRACTION APPARATUS, FACE COMPONENT EXTRACTION METHOD AND RECORDING MEDIUM IN WHICH PROGRAM FOR FACE COMPONENT EXTRACTION METHOD IS STORED

• US 20140140624A1
• Filed: 10/10/2013
• Published: 05/22/2014
• Est. Priority Date: 11/21/2012
• Status: Active Grant

• Current Assignee: Casio Computer Company Limited
• Original Assignee: Casio Computer Company Limited
• Inventors: KASAHARA, Hirokiyo, SHIMADA, Keisuke
• Granted Patent: US 9,323,981 B2
• Time in Patent Office: Days
• US Class Current: 382/195
• 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

Abstract

Disclosed is a face component extraction apparatus including an eye detection unit which detects a plurality of combinations of eye regions, each combination forming a pair, a first calculation unit which calculates a first evaluation value for each pair of eye regions, a fitting unit which fits a plurality of extraction models for extracting a plurality of face components in the image based on a number of pairs of eye regions whose first evaluation values are equal to or greater than a predetermined value, a second calculation unit which calculates a second evaluation value for each of a number of pairs of eye regions, and a deciding unit which decides a fitting mode of the plurality of extraction models to be fitted by the fitting unit based on calculation results of a number of second evaluation values by the second calculation unit.

Claims

1. A face component extraction apparatus, comprising:

an eye detection unit which detects a plurality of combinations of eye regions, each combination forming a pair, in a face region in an image;

a first calculation unit which calculates a first evaluation value for each pair of eye regions detected by the eye detection unit, the first calculation unit being related to likeness to eyes;

a fitting unit which fits a plurality of extraction models for extracting a plurality of face components in the image based on a number of pairs of eye regions whose first evaluation values calculated by the first calculation unit are equal to or greater than a predetermined value;

a second calculation unit which calculates a second evaluation value for each of a number of pairs of eye regions, the second evaluation value being related to a fitting condition of the plurality of extraction models which are fitted by the fitting unit; and

a deciding unit which decides a fitting mode of the plurality of extraction models to be fitted by the fitting unit based on calculation results of a number of second evaluation values by the second calculation unit.

2. The face component extraction apparatus as claimed in claim 1, wherein

the deciding unit specifies a highest second evaluation value among the number of second evaluation values which are calculated by the second calculation unit and decides the fitting mode of the plurality of extraction models corresponding to the highest second evaluation value.

3. The face component extraction apparatus as claimed in claim 1, wherein

the fitting unit specifies a number of pairs of eye regions whose evaluation values are equal to or greater than a predetermined value by setting a highest first evaluation value among the plurality of first evaluation values calculated by the first calculation unit as a reference and fits the plurality of extraction models in the image based on each of the pairs of eye regions.

4. The face component extraction apparatus as claimed in claim 1 further comprising:

a face detection unit which detects a face region in the image, wherein

the eye detection unit detects a pair of eye regions corresponding to a pair of eyes in the face region detected by the face detection unit.

5. The face component extraction apparatus as claimed in claim 4, wherein

the fitting unit fits the plurality of extraction models in the image based on a detection result of the face region by the face detection unit.

6. The face component extraction apparatus as claimed in claim 1 further comprising:

a first creating unit which creates a face component image relating to the plurality of face components based on the fitting mode of the plurality of extraction models decided by the deciding unit; and

a second creating unit which creates a portrait image of the image by using the face component image generated by the first creating unit.

7. A face component extraction method using a face component extraction apparatus, comprising:

detecting a plurality of combinations of eye regions, each combination forming a pair, in a face region in an image;

first calculating of a first evaluation value for each pair of eye regions detected by the eye detection unit, the first calculation unit being related to likeness to eyes;

fitting a plurality of extraction models for extracting a plurality of face components in the image based on a number of pairs of eye regions whose first evaluation values calculated in the first calculating are equal to or greater than a predetermined value;

second calculating of a second evaluation value for each of a number of pairs of eye regions, the second evaluation value being related to a fitting condition of the plurality of extraction models which are fitted in the fitting; and

deciding a fitting mode of the plurality of extraction models to be fitted in the fitting based on calculation results of a number of second evaluation values calculated in the second calculating.

8. A non-transitory computer readable recording medium having a program stored thereon that can be read by a computer in a face component extraction apparatus, the program making the computer function as:

an eye detection unit which detects a plurality of combinations of eye regions, each combination forming a pair, in a face region in an image;

a first calculation unit which calculates a first evaluation value for each pair of eye regions detected by the eye detection unit, the first calculation unit being related to likeness to eyes;

a fitting unit which fits a plurality of extraction models for extracting a plurality of face components in the image based on a number of pairs of eye regions whose first evaluation values calculated by the first calculation unit are equal to or greater than a predetermined value;

a second calculation unit which calculates a second evaluation value for each of a number of pairs of eye regions, the second evaluation value being related to a fitting condition of the plurality of extraction models which are fitted by the fitting unit; and

a deciding unit which decides a fitting mode of the plurality of extraction models to be fitted by the fitting unit based on calculation results of a number of second evaluation values by the second calculation unit.