Patents by Inventor Brendan J. Frey
Brendan J. Frey has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20120185172Abstract: A method, system and apparatus for processing requests is provided. The method comprises storing, in a memory, splicing code data comprising a plurality of features, the splicing code data further comprising, in association with each feature, at least one parameter defining the activity of the feature in splicing regulation; receiving a request at a processor, the request identifying at least a first portion of a genomic sequence; receiving, at the processor, at least a second portion of the genomic sequence that is relevant to the first portion; generating, at the processor, a feature set comprising at least one feature from the second portion of the genomic sequence; generating, based on the splicing code data and the feature set, a response to the request, the response comprising at least one of predicted changes in inclusion levels and a predicted feature map for at least one condition; and transmitting the response.Type: ApplicationFiled: April 29, 2011Publication date: July 19, 2012Inventors: Joseph BARASH, Brendan J. FREY, Benjamin J. BLENCOWE
-
Patent number: 7940264Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: GrantFiled: June 6, 2010Date of Patent: May 10, 2011Assignee: Microsoft CorporationInventors: Nebojsa Jojic, Brendan J. Frey
-
Publication number: 20100238266Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: ApplicationFiled: June 6, 2010Publication date: September 23, 2010Applicant: MICROSOFT CORPORATIONInventors: Nebojsa Jojic, Brendan J. Frey
-
Patent number: 7750903Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: GrantFiled: September 23, 2006Date of Patent: July 6, 2010Assignee: Microsoft CorporationInventors: Nebojsa Jojic, Brendan J. Frey
-
Patent number: 7750904Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: GrantFiled: September 23, 2006Date of Patent: July 6, 2010Assignee: Microsoft CorporationInventors: Nebojsa Jojic, Brendan J. Frey
-
Patent number: 7702489Abstract: The present invention provides a method of constructing recognition models. Under the method, a set of probabilities is estimated for values of a hidden variable. A Fourier transform is determined for the set of probabilities and is used to determine a Fourier transform of an estimated prototype pattern. The inverse Fourier transform is then determined for the Fourier transform of the estimated prototype pattern to form an estimated prototype pattern.Type: GrantFiled: November 1, 2002Date of Patent: April 20, 2010Assignee: Microsoft CorporationInventors: Nebojsa Jojic, Brendan J. Frey
-
Patent number: 7680353Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: GrantFiled: September 23, 2006Date of Patent: March 16, 2010Assignee: Microsoft CorporationInventors: Nebojsa Jojic, Brendan J. Frey
-
Patent number: 7451083Abstract: A method and computer-readable medium are provided for identifying clean signal feature vectors from noisy signal feature vectors. One aspect of the invention includes using an iterative approach to identify the clean signal feature vector. Another aspect of the invention includes using the variance of a set of noise feature vectors and/or channel distortion feature vectors when identifying the clean signal feature vectors.Type: GrantFiled: July 20, 2005Date of Patent: November 11, 2008Assignee: Microsoft CorporationInventors: Brendan J. Frey, Alejandro Acero, Li Deng
-
Patent number: 7310599Abstract: A method and computer-readable medium are provided for identifying clean signal feature vectors from noisy signal feature vectors. Aspects of the invention use mixtures of distributions of noise feature vectors and/or channel distortion feature vectors when identifying the clean signal feature vectors.Type: GrantFiled: July 20, 2005Date of Patent: December 18, 2007Assignee: Microsoft CorporationInventors: Brendan J. Frey, Alejandro Acero, Li Deng
-
Patent number: 7113185Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: GrantFiled: November 14, 2002Date of Patent: September 26, 2006Assignee: Microsoft CorporationInventors: Nebojsa Jojic, Brendan J. Frey
-
Patent number: 6985858Abstract: A method and computer-readable medium are provided for identifying clean signal feature vectors from noisy signal feature vectors. The method is based on variational inference techniques. One aspect of the invention includes using an iterative approach to identify the clean signal feature vector. Another aspect of the invention includes using the variance of a set of noise feature vectors and/or channel distortion feature vectors when identifying the clean signal feature vectors. Further aspects of the invention use mixtures of distributions of noise feature vectors and/or channel distortion feature vectors when identifying the clean signal feature vectors. Additional aspects of the invention include using a variance for the noisy signal feature vector conditioned on fixed values of noise, channel transfer function, and clean speech, when identifying the clean signal feature vector.Type: GrantFiled: March 20, 2001Date of Patent: January 10, 2006Assignee: Microsoft CorporationInventors: Brendan J. Frey, Alejandro Acero, Li Deng
-
Publication number: 20040095374Abstract: A simplified general model and an associated estimation algorithm is provided for modeling visual data such as a video sequence. Specifically, images or frames in a video sequence are represented as collections of flat moving objects that change their appearance and shape over time, and can occlude each other over time. A statistical generative model is defined for generating such visual data where parameters such as appearance bit maps and noise, shape bit-maps and variability in shape, etc., are known. Further, when unknown, these parameters are estimated from visual data without prior pre-processing by using a maximization algorithm. By parameter estimation and inference in the model, visual data is segmented into components which facilitates sophisticated applications in video or image editing, such as, for example, object removal or insertion, tracking and visual surveillance, video browsing, photo organization, video compositing, etc.Type: ApplicationFiled: November 14, 2002Publication date: May 20, 2004Inventors: Nebojsa Jojic, Brendan J. Frey
-
Publication number: 20040088272Abstract: The present invention provides a method of constructing recognition models. Under the method, a set of probabilities is estimated for values of a hidden variable. A Fourier transform is determined for the set of probabilities and is used to determine a Fourier transform of an estimated prototype pattern. The inverse Fourier transform is then determined for the Fourier transform of the estimated prototype pattern to form an estimated prototype pattern.Type: ApplicationFiled: November 1, 2002Publication date: May 6, 2004Inventors: Nebojsa Jojic, Brendan J. Frey
-
Publication number: 20020173953Abstract: A method and computer-readable medium are provided for identifying clean signal feature vectors from noisy signal feature vectors. The method is based on variational inference techniques. One aspect of the invention includes using an iterative approach to identify the clean signal feature vector. Another aspect of the invention includes using the variance of a set of noise feature vectors and/or channel distortion feature vectors when identifying the clean signal feature vectors. Further aspects of the invention use mixtures of distributions of noise feature vectors and/or channel distortion feature vectors when identifying the clean signal feature vectors. Additional aspects of the invention include using a variance for the noisy signal feature vector conditioned on fixed values of noise, channel transfer function, and clean speech, when identifying the clean signal feature vector.Type: ApplicationFiled: March 20, 2001Publication date: November 21, 2002Inventors: Brendan J. Frey, Alejandro Acero, Li Deng