Abstract: Features are extracted from a test and reference image to generate a test and reference record. Each feature has a location, and orientation, and furthermore, the features of the reference records also have associated weights. The features of the test record are approximately aligned with the features of the reference record. Then, differences between the locations and orientations of the features of the reference record and the features of the test record are measured, and the weights of all features of the reference record that are less than a predetermined difference when compared with the features of the test record are summed to determine a similarity score that the test record matches the reference record.

Abstract: A method recognizes a face in an image. A morphable model having shape and pose parameters is fitted to a face in an image to construct a three-dimensional model of the face. Texture is extracted from the face in the image using the three-dimensional model. The shape and texture are projected into a bilinear illumination model to generate illumination bases for the face in the image. The illumination bases for the face in the image are compared to illumination bases of each of a plurality of bilinear illumination models of known faces to identify the face in the image.

Abstract: A method generates a three-dimensional, bi-linear, illumination model for arbitrary faces. A large number of images are acquired of many different faces. For each face, multiple images are acquired with varying poses and varying illumination. A three-mode singular value decomposition is applied to the images to determine parameters of the model. The model can be fit to a probe image of an unknown face. Then, the model can be compared with models of a gallery of images of unknown faces to recognize the face in the probe image.

Abstract: A method determines an optimal set of viewpoints to acquire a 3D shape of a face. A view-sphere is tessellated with a plurality of viewpoint cells. The face is at an approximate center of the view-sphere. Selected viewpoint cells are discarded. The remaining viewpoint cells are clustered to a predetermined number of viewpoint cells according to a silhouette difference metric. The predetermined number of viewpoint cells are searched for a set of optimal viewpoint cells to construct a 3D model of the face.

Abstract: A method determines a direction of a principal light source in an image. An input image I is acquired of an object illuminated by a principal light source from an unknown direction. The input image includes an array of pixels, and each pixel has an intensity I(x, y). An intensity vector {right arrow over (I)} is constructed from the intensities of the input image. An albedo map ? is defined for the object. An albedo vector {right arrow over (?)} is constructed from the albedo map. A shape matrix N is generated for the object. The albedo vector {right arrow over (?)} is multiplied by the shape matrix N to obtain a shape-albedo matrix A. Then, a direction s* to the principal light source is estimated from the intensity vector {right arrow over (I)}, the albedo vector {right arrow over (?)} and the shape-albedo matrix A according to an optimization s * = arg ? ? min s ? ? I ? - ? ? - As ? .

Abstract: A method maximizes a candidate solution to a cardinality-constrained combinatorial optimization problem of sparse principal component analysis. An approximate method has as input a covariance matrix A, a candidate solution, and a sparsity parameter k. A variational renormalization for the candidate solution vector x with regards to the eigenvalue structure of the covariance matrix A and the sparsity parameter k is then performed by means of a sub-matrix eigenvalue decomposition of A to obtain a variance maximized k-sparse eigenvector x that is the best possible solution. Another method solves the problem by means of a nested greedy search technique that includes a forward and backward pass. An exact solution to the problem initializes a branch-and-bound search with an output of a greedy solution.

Abstract: A computer implemented method maximizes candidate solutions to a cardinality-constrained combinatorial optimization problem of sparse linear discriminant analysis. A candidate sparse solution vector x with k non-zero elements is inputted, along with a pair of covariance matrices A, B measuring between-class and within-class covariance of binary input data to be classified, the sparsity parameter k denoting a desired cardinality of a final solution vector. A variational renormalization of the candidate solution vector x is performed with regards to the pair of covariance matrices A, B and the sparsity parameter k to obtain a variance maximized discriminant eigenvector {circumflex over (x)} with cardinality k that is locally optimal for the sparsity parameter k and zero-pattern of the candidate sparse solution vector x, and is the final solution vector for the sparse linear discriminant analysis optimization problem.

Abstract: A method reconstructs or synthesizes heads from 3D models of heads and 2D silhouettes of heads. A 3D statistical model is generated from multiple real human heads. The 3D statistical model includes a model parameter in the form of basis vectors and corresponding coefficients. Multiple 2D silhouettes of a particular head are acquired using a camera for example. The 3D statistical model is fitted to multiple 2D silhouettes to determine a particular value of the model parameter corresponding to the plurality of 2D silhouettes. Then, the 3D statistical model is rendered according to the particular value of the model parameter to reconstruct the particular head.

Abstract: A method generates a three-dimensional, bi-linear, illumination model for arbitrary faces. A large number of images are acquired of many different faces. For each face, multiple images are acquired with varying poses and varying illumination. A three-mode singular value decomposition is applied to the images to determine parameters of the model. The model can be fit to a probe image of an unknown face. Then, the model can be compared with models of a gallery of images of unknown faces to recognize the face in the probe image.

Abstract: A method recognizes a face in an image. A morphable model having shape and pose parameters is fitted to a face in an image to construct a three-dimensional model of the face. Texture is extracted from the face in the image using the three-dimensional model. The shape and texture are projected into a bilinear illumination model to generate illumination bases for the face in the image. The illumination bases for the face in the image are compared to illumination bases of each of a plurality of bilinear illumination models of known faces to identify the face in the image.

Abstract: A method determines a direction of a principal light source in an image. An input image I is acquired of an object illuminated by a principal light source from an unknown direction. The input image includes an array of pixels, and each pixel has an intensity I(x, y). An intensity vector {right arrow over (I)} is constructed from the intensities of the input image. An albedo map ? is defined for the object. An albedo vector {right arrow over (?)} is constructed from the albedo map. A shape matrix N is generated for the object. The albedo vector {right arrow over (?)} is multiplied by the shape matrix N to obtain a shape-albedo matrix A. Then, a direction s* to the principal light source is estimated from the intensity vector {right arrow over (I)}, the albedo vector {right arrow over (?)} and the shape-albedo matrix A according to an optimization s * = arg ? ? ? min s ? ? I ? - ? ? - As ? .

Abstract: A method learns a binary classifier for classifying samples into a first class and a second class. First, a set of training samples is acquired. Each training sample is labeled as either belonging to the first class or to the second class. Pairs of dyadic samples are connected by projection vectors such that a first sample of each dyadic pair belonging to the first class and a second sample of each dyadic pair belonging to the second class. A set of hyperplanes are formed so that the hyperplanes have a surface normal to the projection vectors. One hyperplane from the set of hyperplanes is selected that minimizes a weighted classification error. The set of training samples is then weighted according to a classification by the selected hyperplane. The selected hyperplanes are combined into a binary classifier, and the selecting, weighting, and combining are repeated a predetermined number of iterations to obtain a final classifier for classifying test samples into the first and second classes.

Abstract: A method classifies images of faces according to gender. Training images of male and female faces are supplied to a vector support machine. A small number of support vectors are determined from the training images. The support vectors identify a hyperplane. After training, a test image is supplied to the support vector machine. The test image is classified according to the gender of the test image with respect to the hyperplane.

Abstract: A method reconstructs or synthesizes heads from 3D models of heads and 2D silhouettes of heads. A 3D statistical model is generated from multiple real human heads. The 3D statistical model includes a model parameter in the form of basis vectors and corresponding coefficients. Multiple 2D silhouettes of a particular head are acquired using a camera for example. The 3D statistical model is fitted to multiple 2D silhouettes to determine a particular value of the model parameter corresponding to the plurality of 2D silhouettes. Then, the 3D statistical model is rendered according to the particular value of the model parameter to reconstruct the particular head.

Abstract: A method determines an optimal set of viewpoints to acquire a 3D shape of a face. A view-sphere is tessellated with a plurality of viewpoint cells. The face is at an approximate center of the view-sphere. Selected viewpoint cells are discarded. The remaining viewpoint cells are clustered to a predetermined number of viewpoint cells according to a silhouette difference metric. The predetermined number of viewpoint cells are searched for a set of optimal viewpoint cells to construct a 3D model of the face.

Abstract: A method for visualizing multimedia objects assigns a feature vector to each multimedia object. The feature vector of each multimedia object is reduced to a location vector having a dimensionality of a display device. A cost function is evaluated to determine an optimal location vector for each multimedia object, and each multimedia object is displayed on a display device according to the optimal location vector. The reducing can use principle component analysis. In addition, a relevance score can be determined for each displayed multimedia object, and the multimedia objects can than be visually enhanced according to the relevance score.

Abstract: A method classifies images of faces according to gender. Training images of male and female faces are supplied to a vector support machine. A small number of support vectors are determined from the training images. The support vectors identify a hyperplane. After training, a test image is supplied to the support vector machine. The test image is classified according to the gender of the test image with respect to the hyperplane.

Abstract: A method learns a binary classifier for classifying samples into a first class and a second class. First, a set of training samples is acquired. Each training sample is labeled as either belonging to the first class or to the second class. Pairs of dyadic samples are connected by projection vectors such that a first sample of each dyadic pair belonging to the first class and a second sample of each dyadic pair belonging to the second class. A set of hyperplanes are formed so that the hyperplanes have a surface normal to the projection vectors. One hyperplane from the set of hyperplanes is selected that minimizes a weighted classification error. The set of training samples is then weighted according to a classification by the selected hyperplane. The selected hyperplanes are combined into a binary classifier, and the selecting, weighting, and combining are repeated a predetermined number of iterations to obtain a final classifier for classifying test samples into the first and second classes.

Abstract: A method for representing an image in an image retrieval database first separates and filters images to extract color and texture features. The color and texture features of each image are partitioned into a plurality of blocks. A joint distribution of the color features and a joint distribution of the texture features are estimated for each block. The estimated joint distributions are stored in the database with each image to enable retrieval of the images by comparing the estimated joint distributions.

Abstract: Features are extracted from a test and reference image to generate a test and reference record. Each feature has a location, and orientation, and furthermore, the features of the reference records also have associated weights. The features of the test record are approximately aligned with the features of the reference record. Then, differences between the locations and orientations of the features of the reference record and the features of the test record are measured, and the weights of all features of the reference record that are less than a predetermined difference when compared with the features of the test record are summed to determine a similarity score that the test record matches the reference record.