Method of virtually trying on eyeglasses
A method of virtually trying on eyeglasses includes capturing a plurality of images of a user's face, obtaining locations of a plurality of feature points on the user's face in the plurality of images, and using the locations of the plurality of feature points in the plurality of images to create a standard three-dimensional model of the user. Next, a selection is received from the user of a pair of virtual eyeglasses, the selected pair of virtual eyeglasses having a corresponding three-dimensional model depicting the size and shape of the selected pair of virtual eyeglasses. After this, a modified three-dimensional model of the user having the selected pair of virtual eyeglasses superimposed on the user's face is created according to the standard three-dimensional model of the user and the corresponding three-dimensional model of the selected pair of virtual eyeglasses. The result is then displayed for the user to see.
1. A method of virtually trying on eyeglasses, the method comprising:
capturing a plurality of images of a user's face;
obtaining locations of a plurality of feature points on the user's face in the plurality of images;
using the locations of the plurality of feature points in the plurality of images to create a standard three-dimensional model of the user;
receiving selection from the user of a pair of virtual eyeglasses, the selected pair of virtual eyeglasses having a corresponding three-dimensional model depicting the size and shape of the selected pair of virtual eyeglasses;
creating a modified three-dimensional model of the user having the selected pair of virtual eyeglasses superimposed on the user's face according to the standard three-dimensional model of the user and the corresponding three-dimensional model of the selected pair of virtual eyeglasses; and
displaying the resulting modified three-dimensional model of the user having the selected pair of virtual eyeglasses superimposed on the user's face;
wherein the position of the user's face in image I j is calculated as R j M+t j , where R j represents the rotation of the user's face in image I j and t j represents the horizontal displacement of the user's face in image I j ;
wherein the three-dimensional coordinates of a point P (k) is equal to (P x (k) ,P y (k) ,P z (k) ); and
wherein two-dimensional coordinates of point P j (k) in image I j are calculated according to the equation
(
P
j
,
x
(
k
)
,
P
j
,
y
(
k
)
,
1
)
=
A
(
P
x
(
k
)
′
P
z
(
k
)
′
,
P
y
(
k
)
′
P
z
(
k
)
′
,
1
)
T
,
in which updated coordinates of the coordinates (P x (k) ,P y (k) ,P z (k) ) in image I j are calculated according to the equation (P x (k) ′,P y (k) ′,P z (k) ′) T =R j (P x (k) ,P y (k) ,P z (k) ) T +t j , where A represents known coefficients of a camera capturing the image I j and T indicates a matrix transpose function for reflecting a matrix over its main diagonal.
2. The method of claim 1 , wherein when capturing the plurality of images of a user's face, the user's face is rotated for depicting the user's face with different angles of rotation in the plurality of images.
3. The method of claim 2 , wherein the user's face maintains a same expression as the user's face is rotated.
4. The method of claim 1 , wherein obtaining locations of the plurality of feature points on the user's face in the plurality of images comprises obtaining locations of eight or more feature points on the user's face in the plurality of images.
5. The method of claim 1 , wherein obtaining locations of the plurality of feature points on the user's face in the plurality of images comprises obtaining locations of approximately 95 feature points on the user's face in the plurality of images.
6. The method of claim 1 , wherein positions of the plurality of feature points are selected from a group consisting of outer edges of the eyes, outer edges of the mouth, outer edges of the nose, outer edges of the eyebrows, and outer edges of the face.
7. The method of claim 1 , wherein capturing the plurality of images of the user's face comprises capturing ten or more images of the user's face.
8. The method of claim 1 , wherein the standard three-dimensional model of the user is calculated as M ={P (k) } k=1 n , where M represents the standard three-dimensional model of the user, P represents three-dimensional coordinates of an individual feature point of the plurality of feature points, n represents the total number of the plurality of feature points, and k is an integer incremented from 1 to n.
9. The method of claim 1 , wherein in the modified three-dimensional model of the user having the selected pair of virtual eyeglasses superimposed on the user's face, superimposing comprises hiding a temple of the selected pair of virtual eyeglasses when the temple would be blocked by the user's face as the user's face is rotated.
10. The method of claim 1 , further comprising:
receiving selection from the user of a different pair of virtual eyeglasses, the different pair of virtual eyeglasses having a corresponding different three-dimensional model depicting the size and shape of the different pair of virtual eyeglasses; and
creating an updated modified three-dimensional model of the user having the different pair of virtual eyeglasses superimposed on the user's face according to the standard three-dimensional model of the user and the corresponding different three-dimensional model of the different pair of virtual eyeglasses.