Abstract: An input signal, in the form of a sequence of feature vectors, is transformed to an output signal by first storing parameters of a model of the input signal in a memory. Using the vectors and the parameters, a sequence of vectors of hidden variables is inferred. There is at least one vector hn of hidden variables hi,n for each feature vector xn, and each hidden variable is nonnegative. The output signal is generated using the feature vectors, the vectors of hidden variables, and the parameters. Each feature vector xn is dependent on at least one of the hidden variables hi,n for the same n. The hidden variables are related according to h i , n = ? j , l ? ? c i , j , l ? ? l , n ? h j , n - 1 , where j and l are summation indices. The parameters include non-negative weights ci,j,l, and ?l,n are independent non-negative random variables.

Abstract: Embodiment of the invention describes a system and a method for positioning an object from a first pose to a second pose. The method displays concurrently a first image and a second image on a display device, wherein the first image and the second image represent respectively the first pose and the second pose of the object. The method updates, in response to a change in a pose of the display device, the second image such that an alignment of the first image and the second image on the display device depends on the change in the pose of the display device, and positions the object from the first pose to the second pose based on the change in the pose of the display device.

Abstract: A method and a system for selecting antennas in an orthogonal frequency division multiple access (OFDMA) network. The network includes a base station and mobile stations. The mobile station includes multiple antennas. The base station and the mobile stations communicate with each other using frames. The base station transmits a down-link (DL) subframe of the frame to the mobile station. The DL subframe allocates one or more symbols and one or more subcarriers of an up-link (UL) subframe of the frame for antenna selection signals. The mobile station transmits the UL subframe including the antenna selection signals at the allocated subcarriers and symbols to the base station. Then, the base station selects a subset of the antennas based on the antenna selection signals.

Abstract: A method analyzes a radio signal received via a wireless channel. The radio signal includes multiple frames representing a transmitted symbol. Energy of each frame is sampled during multiple of non-overlapping time windows. The sampled energies are stored in an energy matrix indexed by the number of frames and the number of time windows in each frame to analyze the radio signal.

Abstract: A method allocates radio channel resource in an orthogonal frequency-division multiple access network including a set of base stations (BS) and a set of mobile stations (MS). For each BS, a diversity set is maintained for the sets of MS. Each BS determines possible interference at the MS based on the diversity set. A graph is constructed, in which nodes represent the sets of MS, and each edge between a pair of nodes represents channel interference between the MS represented by the pair of nodes. A weight is assigned to each edge, which reflects interference and signal strength on the subchannel between the two MSs connected by the edge. Channel resources are allocated to the MS based on the graph.

Abstract: A method allocates radio channel resources in an orthogonal frequency-division multiple access network including a set of base stations (BS) and a set of mobile stations (MS). For each BS, a diversity set is maintained for the sets of MS. Each BS determines possible interference at the MS based on the diversity set. A graph is constructed, in which nodes represent the sets of MS, mid each edge between a pair of nodes represents channel interference between the MS represented by the pair of nodes. A weight is assigned to each edge, which reflects interference between the two MSs connected by the edge. The interference graph is partitioned into non-overlapping clusters of nodes based on a structure of the interference graph, the potential interference, so that a sum of the weights of the edges between each cluster is maximized. Based upon the graph partitioning, the channel resources are allocated to the mobile stations in order to maximize the system capacity.

Abstract: A computer implemented method extracts an integral histogram from sampled data, such as time series data, images, and volumetric data. First, a set of samples is acquired from a real-world signal. The set of samples is scanned in a predetermined order. For each current sample, an integral histogram integrating a histogram of the current sample and integral histograms of previously scanned samples is constructed.

Abstract: A method according to the invention separates multiple audio signals recorded as a mixed signal via a single channel. The mixed signal is A/D converted and sampled. A sliding window is applied to the samples to obtain frames. The logarithms of the power spectra of the frames are determined. From the spectra, the a posteriori probabilities of pairs of spectra are determined. The probabilities are used to obtain Fourier spectra for each individual signal in each frame. The invention provides a minimum-mean-squared error metho or a soft mask method for making this determination. The Fourier spectra are inverted to obtain corresponding signals, which are concatenated to recover the individual signals.

Abstract: A method constructs three-dimensional (3D) models of a scene from a set of two-dimensional (2D) input images. The 3D model can then be used to reconstruct the scene from arbitrary viewpoints. A user segments and labels a set of corresponding polygonal regions in each image using conventional photo-editing tools. The invention constructs the model so that the model has a maximum volume that is consistent with the set of labeled regions in the input images. The method according to the invention directly constructs the polygonal model.

Abstract: A method and apparatus determine an antialiased intensity of a component of a pixel. A set of objects is represented by a set of two-dimensional distance fields. For each distance field, a corresponding set of sample points is associated with the component and a corresponding distance is determined using the corresponding set of sample points. The corresponding distances are combined to determine a combined distance that is mapped to the antialiased intensity of the component of the pixel.

Abstract: A method compensates for drift in macroblocks of a partially decoded input bitstream. The macroblocks include intra-mode and inter-mode macroblocks, and each macroblock includes DCT coefficients, and at least one motion vector. An estimate of drift is measured in the partially decoded input bitstream. The estimated drift is translated into an intra refresh rate. The modes of inter-mode macroblock are mapped to intra-mode macroblock according to the refresh rate. The DCT coefficients and the motion vector for each changed macroblock are modified in accordance with the mapping for each changed macroblock.

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 models a graphics object by providing a model of the graphics object and generating a first adaptively sampled distance field for the model. Next, a topological hint is constructed. A second adaptively sampled distance field is generated from the topological hint. First locations of the second adaptively sampled distance field are sampled to determine a corresponding topological feature for each location. Second locations in the first adaptively sampled distance are determined from the corresponding topological features. The first adaptively sampled distance field is then sampled at the second locations to determine a distance value for each of the second locations to model the graphics object according to the topological hint.

Abstract: A method stores a plurality of sequentially ordered data samples in a plurality of sequential storage segments such that one of the plurality of sequentially ordered data samples is stored at one of the plurality of sequential storage segments which is non-sequential to another of the plurality of sequential storage segments in which an immediately preceding one of the plurality of sequentially ordered data samples is stored. Another one of the stored plurality of sequentially ordered data samples, other than a first ordered one of the plurality of sequentially ordered data samples, is overwritten with a further data sample, the further data sample being immediately sequential to a last ordered one of the plurality of sequentially ordered data samples.

Abstract: A chemochromic sensor includes multiple bi-directional LEDs, each optically aligned with one or more test areas. Each LED is used as both a light emitter when driven in forward bias, and as a light detector when driven in reverse bias. By alternating the bias on the LEDs, multi-way light measurements of the test area can be obtained.

Abstract: A method models a non-rigid three-dimensional object directly from a sequence of images. A shape of the object is represented as a matrix of 3D points, and a basis of possible deformations of the object is represented as a matrix of displacements of the 3D points. The matrices of 3D points and displacements forming a model of the object. Evidence for an optical flow is determined from image intensities in a local region near each 3D point. The evidence is factored into 3D rotation, translation, and deformation coefficients of the model to track the object in the video.

Abstract: A method decodes a video encoded as a compressed bitstream including frames constructed as blocks. The bitstream is variable length decoded and inverse quantized to determine motion vectors and DCT coefficients of the blocks of the bitstream. The DCT coefficients are then inverse discrete cosine transformed. A low-resolution block is generated for each reconstructed block, and the reconstructed blocks are motion compensated with the low-resolution blocks to decode the video.

Abstract: A method determines poses of a sensors distributed in an environment. A signal of the environment is acquired by each sensor. Features in each signal that correspond to the features in at least one other signal are identified. Directions between the sensors and the corresponding features are determined. Nullspaces of the directions are used to construct a matrix. A nullspace eigenvector is determined of the matrix, and then the nullspace eigenvector is reconfigured to a matrix specifying the locations of the sensors.

Abstract: A method renders a region of a composite glyph. The composite glyph is defined by a set of elements. A set of two-dimensional distance fields is generated using the set of elements, where each two-dimensional distance field is partitioned into cells and each cell includes a method for reconstructing the corresponding two-dimensional distance field within the cell. A composition of the set of two-dimensional distance fields represents the composite glyph. The region of the composite glyph is then rendered using the set of two-dimensional distance fields.

Abstract: A method describes activity in a video sequence. The method measures intensity, direction, spatial, and temporal attributes in the video sequence, and the measured attributes are combined in a digital descriptor of the activity of the video sequence.