Method and apparatus for expressing rigid area based on expression control points
A method and apparatus for expressing a rigid area based on expression control points that express a motion change of the rigid area with respect to an input expression. The method and apparatus express the rigid area by interpolating a change of expression of an avatar using a weight applied to an expression vector related to a plurality of expression samples to express the input expression and also using a transformation matrix set corresponding to the rigid area of each of the plurality of expression samples.
1. A method for expressing a rigid area corresponding to an input expression, the method comprising:
generating, by a processor, a pose vector to express the input expression;
calculating weights applied to expression vectors related to a plurality of expression samples, based on the generated pose vector, wherein the expression vectors represent the plurality of expression samples expressing a plurality of expressions of an avatar;
calculating a transformation matrix of the rigid area by applying the calculated weights to transformation matrices related to the plurality of expression samples,
wherein each of the transformation matrices indicates a degree of change between expression control points of a rigid area of the corresponding expression sample and expression control points of a rigid area of a basic expression and the transformation matrices are used for expressing the plurality of expression samples; and
expressing the rigid area based on the calculated transformation matrix and the basic expression, wherein the expression vectors and the transformation matrices are stored in an avatar pose database (DB).
2. The method of claim 1 , further comprising:
setting correlations between feature points and expression control points related to the input expression, using vertices of a face mesh related to the input expression and the feature points.
3. The method of claim 1 , wherein the generating of the pose vector generates the pose vector based on expression control points related to the input expression.
4. The method of claim 1 , further comprising:
setting a transformation matrix corresponding to the rigid area of each of the plurality of expression samples.
5. The method of claim 4 , wherein the setting of the transformation matrix sets the transformation matrix containing a 3-dimensional (3D) position, rotation, enlargement and reduction with respect to the rigid area of each of the plurality of expression samples.
6. The method of claim 1 , further comprising:
generating a face mesh to control an avatar using expression control points related to the input expression.
7. The method of claim 1 , wherein the calculating the transformation matrix comprises:
calculating an interpolation function representing correlations between the expression vectors and a position of the rigid area of each of the plurality of expression samples.
8. The method of claim 1 , wherein the rigid area comprises at least one of eyes and a jaw.
9. The method of claim 1 , further comprising:
defining a mapping function between feature points and the expression control points of the rigid area of the corresponding expression sample; and
converting coordinates of the feature points into the expression control points of the rigid area of the corresponding expression sample on a face mesh, by calculating positions of vertices related to a random image frame using the mapping function.
10. A non-transitory computer-readable recording medium storing a program for instructing a computer to implement the method of claim 1 .
11. The method of claim 1 , wherein a motion of the rigid area is expressed using the transformation matrix, without changing a shape of the rigid area.
12. An apparatus for expressing a rigid area corresponding to an input expression, the apparatus comprising:
a processor comprising:
a generation unit to generate a pose vector for expressing the input expression;
a calculation unit to calculate weights applied to expression vectors related to a plurality of expression samples, based on the generated pose vector, wherein the expression vectors represent the plurality of expression samples expressing a plurality of expressions of an avatar;
an interpolation unit to calculate a transformation matrix of the rigid area by applying the calculated weights to transformation matrices related to the plurality of expression samples,
wherein each of the transformation matrices indicates a degree of change between expression control points of a rigid area of the corresponding expression sample and expression control points of a rigid area of a basic expression and the transformation matrices are used for expressing the plurality of expression samples; and
an expressing unit to express the rigid area based on the calculated transformation matrix and the basic expression, wherein the expression vectors and the transformation matrices are stored in an avatar pose database (DB).
13. The apparatus of claim 12 , wherein correlations are set between feature points and expression control points related to the input expression, using vertices of a face mesh related to the input expression and the feature points.
14. The apparatus of claim 12 , wherein the generation unit generates the pose vector based on expression control points related to the input expression.
15. The apparatus of claim 12 , further comprising a set unit to set a transformation matrix corresponding to the rigid area of each of the plurality of expression samples.
16. The apparatus of claim 15 , wherein the set unit sets the transformation matrix containing a 3-dimensional (3D) position, rotation, enlargement and reduction with respect to the rigid area of each of the plurality of expression samples.
17. The apparatus of claim 12 , further comprising a mesh generation unit to generate a face mesh for controlling an avatar using expression control points related to the input expression.
18. The apparatus of claim 12 , wherein the interpolation unit further comprises a function calculator to calculate an interpolation function representing correlations between the expression vectors and a position of the rigid area of each of the plurality of expression samples.
19. The apparatus of claim 12 , wherein the rigid area comprises at least one of eyes and a jaw.
20. The apparatus of claim 12 , wherein the apparatus defines a mapping function between feature points and the expression control points of the rigid area of the corresponding expression sample and converts coordinates of the feature points into the expression control points of the rigid area of the corresponding expression sample on a face mesh, by calculating positions of vertices related to a random image frame using the mapping function.