Facial feature detection on mobile devices
First Claim
1. A method for locating facial features on a facial area, comprising:
- locating an eye location, comprising;
generating an intensity response map by applying a 3-rectangle filter to pixels in an eye slice from the facial area;
applying K-mean clustering to the intensity response map to determine the eye location;
locating an eye corner location, comprising;
preprocessing pixels in an eye patch from the facial area by applying logarithm transform and grayscale stretching to generate a grayscale eye patch;
generating a binary map by binarizing the grayscale eye patch using a threshold based on a histogram of the grayscale eye patch;
estimating the eye corner location by averaging coordinates weighted by minimal eigenvalues of spatial gradient matrices in an eye corner search region based on the binary map;
locating a mouth corner location, comprising;
generating another intensity response map by applying another 3-rectangle filter to pixels in a mouth patch from the facial area;
generating another binary map by binarizing the another intensity response map using another threshold based on another histogram of the another intensity response map;
locating a chin location, comprising;
applying angle constrained gradient analysis to reject locations in a chin search region from the facial area that cannot be the chin location;
locating a cheek location, comprising;
applying another angle constrained gradient analysis to reject locations in a cheek search region from the facial area that cannot be the cheek location;
removing falsely detected cheek location candidates by parabola fitting curves through the chin location and cheek location candidates.
4 Assignments
0 Petitions
Accused Products
Abstract
Locating an eye includes generating an intensity response map by applying a 3-rectangle filter and applying K-mean clustering to the map to determine the eye. Locating an eye corner includes applying logarithm transform and grayscale stretching to generate a grayscale eye patch, generating a binary map of the patch by using a threshold based on a histogram of the patch, and estimating the eye corner by averaging coordinates weighted by minimal eigenvalues of spatial gradient matrices in a search region based on the binary map. Locating a mouth corner includes generating another intensity response map and generating another binary map using another threshold based on another histogram of the intensity response map. Locating a chin or a cheek includes applying angle constrained gradient analysis to reject locations that cannot be the chin or cheek. Locating a cheek further includes removing falsely detected cheeks by parabola fitting curves through the cheeks.
-
Citations
29 Claims
-
1. A method for locating facial features on a facial area, comprising:
-
locating an eye location, comprising;
generating an intensity response map by applying a 3-rectangle filter to pixels in an eye slice from the facial area;
applying K-mean clustering to the intensity response map to determine the eye location;
locating an eye corner location, comprising;
preprocessing pixels in an eye patch from the facial area by applying logarithm transform and grayscale stretching to generate a grayscale eye patch;
generating a binary map by binarizing the grayscale eye patch using a threshold based on a histogram of the grayscale eye patch;
estimating the eye corner location by averaging coordinates weighted by minimal eigenvalues of spatial gradient matrices in an eye corner search region based on the binary map;
locating a mouth corner location, comprising;
generating another intensity response map by applying another 3-rectangle filter to pixels in a mouth patch from the facial area;
generating another binary map by binarizing the another intensity response map using another threshold based on another histogram of the another intensity response map;
locating a chin location, comprising;
applying angle constrained gradient analysis to reject locations in a chin search region from the facial area that cannot be the chin location;
locating a cheek location, comprising;
applying another angle constrained gradient analysis to reject locations in a cheek search region from the facial area that cannot be the cheek location;
removing falsely detected cheek location candidates by parabola fitting curves through the chin location and cheek location candidates. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29)
-
Specification