Three-dimensional face recognition
First Claim
1. Apparatus for processing 3-dimensional data of a geometric body for matching, said geometric body comprising an outer surface having a topography, said apparatus comprising:
- a geodesic converter, configured for receiving 3-dimensional topographical data of said geometric body, and for converting said 3-dimensional topographical data into a series of geodesic distances between pairs of points of said topographical data on a manifold, said geodesic distances being measured over said manifold of said geometric body, anda bending invariant multi-dimensional scaler, connected subsequently to said geodesic converter, for forming a bending-invariant low dimensional Euclidean representation of said series of geodesic distances over said manifold, said bending-invariant low dimensional Euclidean representation providing a bending invariant representation of said geometric body suitable for matching with other geometric shapes.
2 Assignments
0 Petitions
Accused Products
Abstract
Apparatus for obtaining 3-Dimensional data of a geometric body for matching, and particularly for use in facial matching, comprises a three dimensional scanner for obtaining three-dimensional topographical data of the body, a triangulator for receiving or forming said data into a triangulated manifold, a geodesic converter, for converting the triangulated manifold into a series of geodesic distances between pairs of points of the manifold, and a multi-dimensional scaler, for forming a low dimensional Euclidean representation of the series of geodesic distances, to give a bending invariant representation of the geometric body. In one variant, matching is carried out by taking the principle eigenvalues from the representation and plotting as co-ordinates in a feature space. Tilted or different expression versions of the same face tend to form clusters in the feature space allowing for matching. The apparatus preferably uses the fast marching method for the triangulated domain to obtain the geodesic distances.
50 Citations
27 Claims
-
1. Apparatus for processing 3-dimensional data of a geometric body for matching, said geometric body comprising an outer surface having a topography, said apparatus comprising:
-
a geodesic converter, configured for receiving 3-dimensional topographical data of said geometric body, and for converting said 3-dimensional topographical data into a series of geodesic distances between pairs of points of said topographical data on a manifold, said geodesic distances being measured over said manifold of said geometric body, and a bending invariant multi-dimensional scaler, connected subsequently to said geodesic converter, for forming a bending-invariant low dimensional Euclidean representation of said series of geodesic distances over said manifold, said bending-invariant low dimensional Euclidean representation providing a bending invariant representation of said geometric body suitable for matching with other geometric shapes. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A method of image preprocessing of three-dimensional topographical data of a manifold surface having a topography for subsequent classification, the method comprising:
-
providing image preprocessing hardware; and within said image preprocessing hardware; placing said three-dimensional topographical data over a two-dimensional manifold to form said manifold surface having a topography, converting said manifold into a series of geodesic distances between pairs of points over said manifold, said geodesic distances being measured over said topography of said manifold, generating a matrix of geodesic distances over said manifold to selected vertices of said manifold, and using multi-dimensional scaling to reduce said matrix to a bending-invariant canonical representation in a low-dimensional Euclidean space, thereby to provide a representation suitable for subsequent classification. - View Dependent Claims (19, 20, 21)
-
-
22. A method of classifying images of three-dimensional bodies having surface topographies, the method comprising:
-
providing image processing hardware; and within said image processing hardware; obtaining bending invariant representations of said three dimensional bodies as canonical form representations derived from geodesic distances between selected sample points taken from surfaces of said bodies, said geodesic distances being measured over said respective surface topographies of said three-dimensional bodies; from each representation deriving co-ordinates on a feature space, and classifying said bodies according to clustering on said feature space. - View Dependent Claims (23, 24, 25)
-
-
26. Apparatus for processing 3-dimensional data of a geometric body having a three-dimensional surface topography, said processing being for matching with other bodies, said apparatus comprising:
a geodesic converter, configured for receiving 3-dimensional topographical data of said geometric body as a 2-dimensional manifold representing said surface, and for converting said manifold into a series of geodesic distances between pairs of points over said manifold, said geodesic distances being measured over a topography of said manifold, thereby to provide a Euclidian bending invariant representation of said geometric body suitable for matching with other geometric bodies, and wherein said geometric body for matching and said other geometric bodies comprises a face. - View Dependent Claims (27)
Specification