Abstract: The present invention relates to a three-dimensional face image capturing method and comprises the steps of: capturing a face region of a user in a direction from a right chin to a left forehead; capturing the face region in a direction from a left forehead to a left chin; capturing the face region in a direction from a left chin to a right forehead; and capturing the face region in a direction from a right forehead to a center of face.

Abstract: Provided is a face pose correction apparatus for correcting a face pose, using a person-specific frontal model specialized for input three-dimensional (3D) facial data. The face pose correction apparatus may comprise: a generation unit for generating corrected first facial data by performing a 3D rigid transformation of input 3D facial data on the basis of a pre-stored 3D standard facial model; and a correction unit for generating a first subject-specialized frontal model which is a bilateral symmetry model of the first facial data, and correcting the first facial data by performing the 3D rigid transformation of the first facial data on the basis of the generated first subject-specialized frontal model.

Abstract: The present invention relates to a three-dimensional face image capturing method and comprises the steps of: capturing a face region of a user in a direction from a right chin to a left forehead; capturing the face region in a direction from a left forehead to a left chin; capturing the face region in a direction from a left chin to a right forehead; and capturing the face region in a direction from a right forehead to a center of face.

Abstract: Provided is a face pose correction apparatus for correcting a face pose, using a person-specific frontal model specialized for input three-dimensional (3D) facial data. The face pose correction apparatus may comprise: a generation unit for generating corrected first facial data by performing a 3D rigid transformation of input 3D facial data on the basis of a pre-stored 3D standard facial model; and a correction unit for generating a first subject-specialized frontal model which is a bilateral symmetry model of the first facial data, and correcting the first facial data by performing the 3D rigid transformation of the first facial data on the basis of the generated first subject-specialized frontal model.

Abstract: A method of determining a facial expression type may include extracting feature points representing facial characteristics from a facial image of a user; grouping the extracted feature points into a plurality of regions; and determining a facial expression type for each of a plurality of groups using feature points and a facial expression type function corresponding to each of the groups.

Abstract: A method of determining a facial expression type may include extracting feature points representing facial characteristics from a facial image of a user; grouping the extracted feature points into a plurality of regions; and determining a facial expression type for each of a plurality of groups using feature points and a facial expression type function corresponding to each of the groups.

Abstract: A health state determining method is disclosed, which includes receiving a facial image of a user, obtaining at least one user database (DB) image corresponding to the facial image, and determining a health state of the user by comparing the facial image to the at least one user DB image.

Abstract: The present invention relates to a method of detecting a skin-colored area using a variable skin color model. In the method, an initial skin-colored area is extracted from an input image using a skin color model, the initial skin-colored area being repeatedly extracted by adjusting a threshold value, which is a criterion for the selection of a skin-colored area, until a ratio of size of the extracted skin-colored area to size of the entire input image becomes less than a predetermined value. A skin-colored area is extracted by applying the threshold value determined in the first stage to the skin color model, the skin color model being modified by analyzing characteristics of object regions and noise regions of the skin-colored area. The skin-colored area detection method can obtain a high skin-colored object detection rate and a high false alarm rate even under various lighting and background conditions.

Abstract: Provided is a three-dimensional model classification method of classifying constitutions.

Abstract: The present invention relates to a method of detecting a skin-colored area using a variable skin color model. In the method, an initial skin-colored area is extracted from an input image using a skin color model, the initial skin-colored area being repeatedly extracted by adjusting a threshold value, which is a criterion for the selection of a skin-colored area, until a ratio of size of the extracted skin-colored area to size of the entire input image becomes less than a predetermined value. A skin-colored area is extracted by applying the threshold value determined in the first stage to the skin color model, the skin color model being modified by analyzing characteristics of object regions and noise regions of the skin-colored area. The skin-colored area detection method can obtain a high skin-colored object detection rate and a high false alarm rate even under various lighting and background conditions.

Abstract: Provided is a three-dimensional model classification method of classifying constitutions.