Sarnoff Corporation

Thomson Licensing

Rca Thomson Licensing Corp.

Clay Douglas Spence

Lucas Cristobal Parra

Paul Sajda

Craig Fancourt

John C Pearson

Jeffrey Lubin

Michael Wade Hansen

Christopher Vincent Alvino

Bogdan Calin Mihai Matei

Aalbert De Vries

Barbara Hanna

Heidi Aarlien Peterson

Arthur Robert Pope

A computer system that processes mixtures of signals, such as speech and noise sources derived from multiple simultaneous microphone recordings, in order to separate them into their underlying sources. A source separation routine optimizes a filter structure by minimizing cross powers of multiple output channels while enforcing geometric constraints on the filter response. The geometric constraints enforce desired responses for given locations of the underlying sources, based on the assumption that the sources are localized in space.

Geometric source separation signal processing technique

An iterative approach to vector median filtering wherein the resulting median vector need not be a member of the original data set. The iterative vector median filtering allows for fast convergence for complex computations and an output which is approximate to the mean, particularly for small data sets. In addition, a method and system for registering and matching 2.5 normal maps is provided. Registration of two maps is performed by optimally aligning their normals through 2-D warping in the image plane in conjunction with a 3-D rotation of the normals. Once aligned, the average dot-product serves as a matching metric for automatic target recognition (ATR).

Method and system for filtering, registering, and matching 2.5D normal maps

A method and apparatus is disclosed for performing blind source separation using convolutive signal decorrelation. For a first embodiment, the method accumulates a length of input signal (mixed signal) that comprises a plurality of independent signals from independent signal sources. The invention then divides the length of input signal into a plurality of T-length periods (windows) and performs a discrete Fourier transform (DFT) on the signal within each T-length period. Thereafter, estimated cross-correlation values are computed using a plurality of the averaged DFT values. A total number of K cross-correlation values are computed, where each of the K values is averaged over N of the T-length periods. Using the cross-correlation values, a gradient descent process computes the coefficients of a FIR filter that will effectively separate the source signals within the input signal. A second embodiment of the invention is directed to on-line processing of the input signal&#x2014;i.e., processing the signal as soon as it arrives with no storage of the signal data. In particular, an on-line gradient algorithm is provided for application to non-stationary signals and having an adaptive step size in the frequency domain based on second derivatives of the cost function. The on-line separation methodology of this embodiment is characterized as multiple adaptive decorrelation.

Method and system for on-line blind source separation

A method and system for improving the accuracy and timeliness of video metadata by incorporating information related to the motion of the camera as derived from the video imagery itself. Frame-to-frame correspondences are used to accurately estimate changes in camera pose. While the method and system do not require geo-registration, geo-registration results, if available, may be considered in processing the video images and generating improved camera pose estimates.

Method and system for improving video metadata through the use of frame-to-frame correspondences

A method and apparatus is disclosed for performing blind source separation using convolutive signal decorrelation. For a first embodiment, the method accumulates a length of input signal (mixed signal) that includes a plurality of independent signals from independent signal sources. The invention then divides the length of input signal into a plurality of T-length periods (windows) and performs a discrete Fourier transform (DFT) on the, signal within each T-length period. Thereafter, estimated cross-correlation values are computed using a plurality of the averaged DFT values. A total number of K cross-correlation values are computed, where each of the K values is averaged over N of the T-length periods. Using the cross-correlation values, a gradient descent process computes the coefficients of a finite impulse response (FIR) filter that will effectively separate the source signals within the input signal. A second embodiment of the invention is directed to on-line processing of the input signal&#x2014;i.e., processing the signal as soon as it arrives with no storage of the signal data. In particular, an on-line gradient algorithm is provided for application to non-stationary signals and having an adaptive step size in the frequency domain based on second derivatives of the cost function. The on-line separation methodology of this embodiment is characterized as multiple adaptive decorrelation.

An apparatus and a concomitant method for modeling local and non-local information in an image to compute an image probability distribution for the image is disclosed. In one embodiment, such an image probability distribution is determined in an object recognition system.

Method and apparatus for image processing by generating probability distribution of images

A signal processing apparatus and concomitant method for learning and integrating features from multiple resolutions for detecting and/or classifying objects. The signal processing apparatus comprises a hierarchical pyramid of neural networks (HPNN) having a &ldquo;fine-to-coarse&rdquo; structure or a combination of the &ldquo;fine-to-coarse&rdquo; and the &ldquo;coarse-to-fine&rdquo; structures.

Method and apparatus for training a neural network to detect objects in an image

A method and apparatus for training and operating a neural network using gated data. The neural network is a mixture of experts that performs &ldquo;soft&rdquo; partitioning of a network of experts. In a specific embodiment, the technique is used to detect malignancy by analyzing skin surface potential data. In particular, the invention uses certain patient information, such as menstrual cycle information, to &ldquo;gate&rdquo; the expert output data into particular populations, i.e., the network is soft partitioned into the populations. An Expectation-Maximization (EM) routine is used to train the neural network using known patient information, known measured skin potential data and correct diagnosis for the particular training data and patient information. Once trained, the neural network parameters are used in a classifier for predicting breast cancer malignancy when given the patient information and skin potentials of other patients.

Method and apparatus for training and operating a neural network for detecting breast cancer

A method and apparatus that performs blind source separation using convolutive signal decorrelation. More specifically, the method accumulates a length of input signal (mixed signal) that includes a plurality of independent signals from independent signal sources. The invention then divides the length of input signal into a plurality of T-length periods (windows) and performs a discrete Fourier transform (DFT) on the signal within each T-length period. Thereafter, estimated cross-correlation values are computed using a plurality of the averaged DFT values. A total number of K cross-correlation values are computed, where each of the K values is averaged over N of the T-length periods. Using the cross-correlation values, a gradient descent process computes the coefficients of a finite impulse response (FIR) filter that will effectively separate the source signals within the input signal. To achieve an accurate solution, the gradient descent process is constrained in that the time-domain values of the filter coefficients can attain only certain values, i.e., the time-domain filter coefficient values W(.tau.) are constrained within each T-length period to be zero for any time t&gt;Q. In this manner, a unique solution for the FIR filter coefficients is computed and a filter produced using these coefficients will effectively separate the source signals.

Convolutive blind source separation using a multiple decorrelation method

A signal processing apparatus and concomitant method for learning and using fidelity metric as a control mechanism and to process large quantities of fidelity metrics from a visual discrimination measure (VDM) to a manageable subjective image quality ratings. The signal processing apparatus incorporates a VDM and a neural network. The VDM receives input image sequences and generates fidelity metrics, which are received by a neural network. The neural network is trained to learn and use the fidelity metrics as a control mechanism, e.g., to control a video encoder.

Method and apparatus for training a neural network to learn and use

A signal processing apparatus and concomitant method for learning and integrating features from multiple resolutions for detecting and/or classifying objects are presented. Neural networks in a pattern tree structure with tree-structured descriptions of objects in terms of simple sub-patterns, are grown and trained to detect and integrate the sub-patterns. A plurality of objective functions and their approximations are presented to train the neural networks to detect sub-patterns of features of some class of objects. Objective functions for training neural networks to detect objects whose positions in the training data are uncertain and for addressing supervised learning where there are potential errors in the training data are also presented.

Method and apparatus for training a neural network to learn hierarchical

A television signal processing apparatus includes at least one neural network for processing a signal representing an image. The neural network includes a plurality of perceptrons each of which includes circuitry for weighting a plurality of delayed representations of said signal, circuitry for providing sums of weighted signals provided by said weighting circuitry, and circuitry for processing said sums with a sigmoidal transfer function. The neural network also includes circuitry for combining output signals provided by ones of said perceptrons for providing a processed signal.

Growing community of inventors

Princeton Junction, NJ, United States of America

Clay Douglas Spence