Apparatus and method of determining facial expression type

1. A computer-implemented method of determining a facial expression type, the method comprising:

extracting feature points representing facial characteristics from a facial image of a user;

grouping the extracted feature points into a plurality of groups based at least in part on a plurality of regions; and

determining, by at least one processor, a facial expression type for each of the plurality of groups of the extracted feature points by applying the extracted feature points to a facial expression type function corresponding to each of the plurality of groups of the extracted feature points,

wherein the determining of the facial expression type for each of the plurality of groups comprises: receiving an input of age information on the user; calibrating the facial expression type function on a basis of the age information; and determining the facial expression type for each of the plurality of groups using the calibrated facial expression type function, and

wherein the calibrating of the facial expression type function comprises: dividing the age information into predetermined age groups; calculating widths and lengths of the respective regions using the feature points; calculating averages and standard deviations of the widths and lengths of the respective regions in the divided age groups; and calibrating the facial expression type function using the averages and standard deviations.

2. The method of claim 1 , further comprising displaying the determined facial expression type of the user.

3. The method of claim 1 , further comprising generating the facial expression type function, wherein the generating of the facial expression type function comprises receiving facial images of a plurality of users; extracting feature points corresponding to regions of the users from the facial images of the users; generating facial expression types using the feature points corresponding to the regions of the users; representing the generated facial expression types in a normal distribution and classifying the facial expression types according to a preset criterion; and generating the facial expression type function using the classified facial expression types.

4. The method of claim 1 , wherein the plurality of regions comprises at least one of a face, an eye, a forehead and a nose, and the facial expression type for each of the plurality of groups is represented using at least one of a face shape, an eye shape, a forehead shape and a nose shape.

5. The method of claim 1 , wherein the plurality of regions comprises at least one of a face, an eye, a forehead and a nose, and the facial expression type is represented based on widths and lengths of regions of a face shape, an eye shape, a forehead shape and a nose shape.

6. The method of claim 4 , wherein the face shape is divided according to a first predetermined criterion into round, oval, long, square, rectangular, trapezoid and inverted triangular shapes, the eye shape is divided according to a second predetermined criterion into a cow eye, a tiger eye, a deer eye and a hawk eye, the forehead shape is divided according to a third predetermined criterion into a sloping and projecting forehead, a straight and projecting forehead, a sloping and flat forehead and a straight and flat forehead, and the nose shape is divided according to a fourth predetermined criterion into a downturned high nose, an upturned high nose, a downturned flat nose and an upturned flat nose.

7. The method of claim 1 , further comprising combining facial expression types of the respective groups.

8. A computer-implemented method of generating a facial expression type function which generates a facial expression type function, the method comprising:

receiving facial images of a plurality of users;

extracting feature points corresponding to regions of the users from the facial images of the users;

generating facial expression types using the feature points corresponding to the regions of the users;

determining at least one criterion for which the generated facial expression types can be represented in a normal distribution;

classifying the facial expression types based at least on the at least one criterion; and

generating, by at least one processor, a facial expression type function using the classified facial expression types,

wherein the regions comprise at least one of a face, an eye, a forehead and a nose, and the facial expression types are represented using at least one of a face shape, an eye shape, a forehead shape and a nose shape, and

wherein the face shape is divided according to a first predetermined criterion into round, oval, long, square, rectangular, trapezoid and inverted triangular shapes, the eye shape is divided according to a second predetermined criterion into a cow eye, a tiger eye, a deer eye and a hawk eye, the forehead shape is divided according to a third predetermined criterion into a sloping and projecting forehead, a straight and projecting forehead, a sloping and flat forehead and a straight and flat forehead, and the nose shape is divided according to a fourth predetermined criterion into a downturned high nose, an upturned high nose, a downturned flat nose and an upturned flat nose.

9. The method of claim 8 , wherein the facial expression types are determined using a distribution of the feature points for the respective regions.

10. The method of claim 8 , wherein the regions comprise at least one of a face, an eye, a forehead and a nose, and the facial expression types are represented based on widths and lengths of the regions.

11. The method of claim 10 , wherein the at least one criterion comprises the widths and lengths of the regions of the users, and the method further comprises:

segmenting the normal distribution into predetermined sections.

12. A non-transitory computer-readable recording medium recording a program to implement a method of generating a facial expression type function, the method comprising:

receiving facial images of a plurality of users;

extracting feature points corresponding to regions of the users from the facial images of the users;

generating facial expression types using the feature points corresponding to the regions of the users;

determining at least one criterion for which the generated facial expression types can be represented in a normal distribution;

classifying the facial expression types based at least in part on the at least one criterion; and

generating a facial expression type function using the classified facial expression types,

wherein the regions comprise at least one of a face, an eye, a forehead and a nose, and the facial expression types are represented using at least one of a face shape, an eye shape, a forehead shape and a nose shape, and

wherein the face shape is divided according to a first predetermined criterion into round, oval, long, square, rectangular, trapezoid and inverted triangular shapes, the eye shape is divided according to a second predetermined criterion into a cow eye, a tiger eye, a deer eye and a hawk eye, the forehead shape is divided according to a third predetermined criterion into a sloping and projecting forehead, a straight and projecting forehead, a sloping and flat forehead and a straight and flat forehead, and the nose shape is divided according to a fourth predetermined criterion into a downturned high nose, an upturned high nose, a downturned flat nose and an upturned flat nose.

13. An apparatus for determining a facial expression type, the apparatus comprising:

an extraction circuit to extract feature points representing facial characteristics from a facial image of a user;

a grouping circuit to group the extracted feature points into a plurality of groups based at least in part on a plurality of regions; and

a facial expression type determination circuit to determine a facial expression type for each of the plurality of groups by inputting the extracted feature points to a facial expression type function corresponding to each of the plurality of groups,

wherein the facial expression type determination circuit is configured to determine the facial expression type for each of the plurality of groups by: receiving an input of age information on the user; calibrating the facial expression type function on a basis of the age information; and determining the facial expression type for each of the plurality of groups using the calibrated facial expression type function, and

wherein the facial expression type determination circuit is configured to calibrate the facial expression type function by: dividing the age information into predetermined age groups; calculating widths and lengths of the respective regions using the feature points; calculating averages and standard deviations of the widths and lengths of the respective regions in the divided age groups; and calibrating the facial expression type function using the averages and standard deviations.

14. The apparatus of claim 13 , further comprising an input circuit to receive an input of age information on the user, wherein the facial expression type determination circuit calibrates the facial expression type function on a basis of the age information and determines the facial expression type for each of the plurality of groups using the calibrated facial expression type function.

15. An apparatus for generating a facial expression type function, the apparatus comprising:

at least one processor circuit configured to:

receive facial images of a plurality of users;

extract feature points corresponding to regions of the users from the facial images of the users;

generate facial expression types using the feature points corresponding to the regions of the users;

determine at least one criterion for which the generated facial expression types can be represented in a normal distribution;

classify the facial expression types based at least in part on the at least one criterion; and

generate a facial expression type function using the classified facial expression types,

wherein the regions comprise at least one of a face, an eye, a forehead and a nose, and the facial expression types are represented using at least one of a face shape, an eye shape, a forehead shape and a nose shape, and

wherein the face shape is divided according to a first predetermined criterion into round, oval, long, square, rectangular, trapezoid and inverted triangular shapes, the eye shape is divided according to a second predetermined criterion into a cow eye, a tiger eye, a deer eye and a hawk eye, the forehead shape is divided according to a third predetermined criterion into a sloping and projecting forehead, a straight and projecting forehead, a sloping and flat forehead and a straight and flat forehead, and the nose shape is divided according to a fourth predetermined criterion into a downturned high nose, an upturned high nose, a downturned flat nose and an upturned flat nose.