U.S. Patent number 7,643,671, Facial Recognition System and Method.
A facial identification system corrects lighting and pose in images prior to comparison with stored images. A three dimensional image is created from an original two dimensional image by combining the image with shape information. An iterative process is used to adjust the shape in order to match the original two dimensional image. A final image is rendered, with adjustments for lighting and pose, from the shape information.
U.S. Patent number 7,643,683: Generating Image Data Bases for Multifeatured Objects.
A method and system for generating an image database for multifeatured objects. The invention uses one or more three-dimensional representations of multifeatured objects of the same generic type as the object to be identified to generate as many images as desired. These images are used to populate an image database that may be used for training and/or identification purposes.
U.S. Patent number 7,643,685: Viewpoint -Invariant Image Matching and Generation of Three-Dimensional Models from Two-Dimensional Imagery.
A method and system for characterizing features in a source multifeatured three-dimensional object, and for locating a best-matching three-dimensional object from a reference database of such objects by performing a viewpoint invariant search among the reference objects. The invention further includes the creation of a three-dimensional representation of the source object by deforming a reference object.
U.S. Patent number 7,853,085. Viewpoint Invariant Detection and Identification of 3D Object from 2D Imagery.
A method and system for detecting a source multifeatured three-dimensional object from two-dimensional source imagery, and for locating a best-matching three-dimensional object from a candidate database of such objects by performing a viewpoint and lighting invariant search among the candidate objects. The invention further includes the method and system for determining the optimally-fitting viewpoint, lighting, and deformation of the located three-dimensional candidate and basing an identification of the candidate with the source on the quality of fit between projections of the candidate and the source imagery.