Patents by Inventor Aharon Bar-Hillel
Aharon Bar-Hillel 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).
-
Patent number: 10488939Abstract: A gesture recognition method comprises receiving at a processor from a sensor a sequence of captured signal frames for extracting hand pose information for a hand and using at least one trained predictor executed on the processor to extract hand pose information from the received signal frames. For at least one defined gesture, defined as a time sequence comprising hand poses, with each of the hand poses defined as a conjunction or disjunction of qualitative propositions relating to interest points on the hand, truth values are computed for the qualitative propositions using the hand pose information extracted from the received signal frames, and execution of the gesture is tracked, by using the truth values to determine which of the hand poses in the time sequence have already been executed and which of the hand poses in the time sequence is expected next.Type: GrantFiled: August 7, 2017Date of Patent: November 26, 2019Assignee: Microsoft Technology Licensing, LLCInventors: Kfir Karmon, Aharon Bar-Hillel, Eyal Krupka, Noam Bloom, Ilya Gurvich, Aviv Hurvitz, Ido Leichter, Yoni Smolin, Yuval Tzairi, Alon Vinnikov
-
Patent number: 10460201Abstract: A computer implemented method of training an image classifier, comprising: receiving training images data labeled according to image classes; selecting reference points of the images; and constructing a set of voting convolutional tables and binary features on a patch surrounding each reference point by performing, for each calculation stage: creating a voting table by: creating first candidate binary features; calculating a global loss reduction for each first candidate binary feature; selecting one first candidate binary feature having minimal global loss reduction; and repeating to select stage-size binary features; and performing a tree split using the voting table by: creating second candidate binary features; calculating a combined loss reduction for each stage-split size group of the second candidate binary features; selecting one of the groups having a maximal combined loss reduction; and creating a child-directing table using the selected binary features.Type: GrantFiled: December 31, 2015Date of Patent: October 29, 2019Assignee: Microsoft Technology Licensing, LLCInventors: Eyal Krupka, Aharon Bar Hillel
-
Publication number: 20180307319Abstract: A gesture recognition method comprises receiving at a processor from a sensor a sequence of captured signal frames for extracting hand pose information for a hand and using at least one trained predictor executed on the processor to extract hand pose information from the received signal frames. For at least one defined gesture, defined as a time sequence comprising hand poses, with each of the hand poses defined as a conjunction or disjunction of qualitative propositions relating to interest points on the hand, truth values are computed for the qualitative propositions using the hand pose information extracted from the received signal frames, and execution of the gesture is tracked, by using the truth values to determine which of the hand poses in the time sequence have already been executed and which of the hand poses in the time sequence is expected next.Type: ApplicationFiled: August 7, 2017Publication date: October 25, 2018Inventors: Kfir KARMON, Eyal KRUPKA, Noam BLOOM, Ilya GURVICH, Aviv HURVITZ, Ido LEICHTER, Yoni SMOLIN, Yuval TZAIRI, Alon VINNIKOV, Aharon BAR-HILLEL
-
Publication number: 20170193328Abstract: A computer implemented method of training an image classifier, comprising: receiving training images data labeled according to image classes; selecting reference points of the images; and constructing a set of voting convolutional tables and binary features on a patch surrounding each reference point by performing, for each calculation stage: creating a voting table by: creating first candidate binary features; calculating a global loss reduction for each first candidate binary feature; selecting one first candidate binary feature having minimal global loss reduction; and repeating to select stage-size binary features; and performing a tree split using the voting table by: creating second candidate binary features; calculating a combined loss reduction for each stage-split size group of the second candidate binary features; selecting one of the groups having a maximal combined loss reduction; and creating a child-directing table using the selected binary features.Type: ApplicationFiled: December 31, 2015Publication date: July 6, 2017Inventors: Eyal KRUPKA, Aharon BAR HILLEL
-
Patent number: 9400945Abstract: A system and method may compare an image vector representing an image feature of a first image fragment of an image to database vectors representing the image feature of database image fragments of database images. It may be determined based on the comparison a first matching database vector of the database vectors which most closely, among the database vectors, describes the first image feature represented by the image vector. The system or method may determine, using a data structure in conjunction with the first matching database vector and previously matched database vectors, a second of the database vectors which includes the first matching database vector and the previously matched database vectors and most closely describes a second image fragment including the first image fragment. The system or method may determine an object feature based on the second database vector.Type: GrantFiled: September 23, 2011Date of Patent: July 26, 2016Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLCInventors: Aharon Bar Hillel, Dan Levi
-
Patent number: 9037520Abstract: A computer-implemented method is provided for statistical data learning under privacy constraints. The method includes: receiving, by a processor, a plurality of pieces of statistical information relating to a statistical object and aggregating, by the processor, the plurality of pieces of statistical information so as to provide an estimation of the statistical object. Each piece of statistical information includes an uncertainty variable, the uncertainty variable being a value determined from a function having a predetermined mean. The number of pieces of statistical information aggregated is proportional to the reliability of the estimation of the statistical object.Type: GrantFiled: November 1, 2012Date of Patent: May 19, 2015Assignee: GM GLOBAL TECHNOLOGY OPERATIONS, LLC.Inventors: Aharon Bar Hillel, Ron M. Hecht, Nadav Lavi
-
Patent number: 8756174Abstract: In one embodiment, the present invention includes a method for training a Support Vector Machine (SVM) on a subset of features (d?) of a feature set having (d) features of a plurality of training instances to obtain a weight per instance, approximating a quality for the d features of the feature set using the weight per instance, ranking the d features of the feature set based on the approximated quality, and selecting a subset (q) of the features of the feature set based on the ranked approximated quality. Other embodiments are described and claimed.Type: GrantFiled: December 22, 2011Date of Patent: June 17, 2014Assignee: Intel CorporationInventors: Eyal Krupka, Aharon Bar-Hillel
-
Patent number: 8724890Abstract: A method is provided for training and using an object classifier to identify a class object from a captured image. A plurality of still images is obtained from training data and a feature generation technique is applied to the plurality of still images for identifying candidate features from each respective image. A subset of features is selected from the candidate features using a similarity comparison technique. Identifying candidate features and selecting a subset of features is iteratively repeated a predetermined number of times for generating a trained object classifier. An image is captured from an image capture device. Features are classified in the captured image using the trained object classifier. A determination is made whether the image contains a class object based on the trained object classifier associating an identified feature in the image with the class object.Type: GrantFiled: April 6, 2011Date of Patent: May 13, 2014Assignee: GM Global Technology Operations LLCInventors: Dan Levi, Aharon Bar Hillel
-
Publication number: 20140122385Abstract: A computer-implemented method is provided for statistical data learning under privacy constraints. The method includes: receiving, by a processor, a plurality of pieces of statistical information relating to a statistical object and aggregating, by the processor, the plurality of pieces of statistical information so as to provide an estimation of the statistical object. Each piece of statistical information includes an uncertainty variable, the uncertainty variable being a value determined from a function having a predetermined mean. The number of pieces of statistical information aggregated is proportional to the reliability of the estimation of the statistical object.Type: ApplicationFiled: November 1, 2012Publication date: May 1, 2014Applicant: General Motors LLCInventors: Aharon Bar Hillel, Ron M. Hecht, Nadav Lavi
-
Patent number: 8630483Abstract: Complex-object detection using a cascade of classifiers for identifying complex-objects parts in an image in which successive classifiers process pixel patches on condition that respective discriminatory features sets of previous classifiers have been identified and selecting additional pixel patches from a query image by applying known positional relationships between an identified complex-object part and another part to be identified.Type: GrantFiled: June 12, 2012Date of Patent: January 14, 2014Assignee: GM Global Technology Operations LLCInventors: Dan Levi, Aharon Bar Hillel
-
Publication number: 20130329988Abstract: Complex-object detection using a cascade of classifiers for identifying complex-objects parts in an image in which successive classifiers process pixel patches on condition that respective discriminatory features sets of previous classifiers have been identified and selecting additional pixel patches from a query image by applying known positional relationships between an identified complex-object part and another part to be identified.Type: ApplicationFiled: June 12, 2012Publication date: December 12, 2013Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLCInventors: Dan LEVI, AHARON BAR HILLEL
-
Publication number: 20130279808Abstract: Complex-object detection using a cascade of classifiers for identifying complex-objects parts in an image in which successive classifiers process pixel patches on condition that respective discriminatory features sets of previous classifiers have been identified and selecting additional pixel patches from a query image by on the basis of probability data.Type: ApplicationFiled: April 20, 2012Publication date: October 24, 2013Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLCInventors: Dan LEVI, Aharon Bar Hillel
-
Publication number: 20130077873Abstract: A system and method may compare an image vector representing an image feature of a first image fragment of an image to database vectors representing the image feature of database image fragments of database images. It may be determined based on the comparison a first matching database vector of the database vectors which most closely, among the database vectors, describes the first image feature represented by the image vector. The system or method may determine, using a data structure in conjunction with the first matching database vector and previously matched database vectors, a second of the database vectors which includes the first matching database vector and the previously matched database vectors and most closely describes a second image fragment including the first image fragment. The system or method may determine an object feature based on the second database vector.Type: ApplicationFiled: September 23, 2011Publication date: March 28, 2013Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLCInventors: Aharon BAR HILLEL, Dan Levi
-
Publication number: 20120257819Abstract: A method is provided for training and using an object classifier to identify a class object from a captured image. A plurality of still images is obtained from training data and a feature generation technique is applied to the plurality of still images for identifying candidate features from each respective image. A subset of features is selected from the candidate features using a similarity comparison technique. Identifying candidate features and selecting a subset of features is iteratively repeated a predetermined number of times for generating a trained object classifier. An image is captured from an image capture device. Features are classified in the captured image using the trained object classifier. A determination is made whether the image contains a class object based on the trained object classifier associating an identified feature in the image with the class object.Type: ApplicationFiled: April 6, 2011Publication date: October 11, 2012Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLCInventors: Dan Levi, Aharon Bar Hillel
-
Publication number: 20120095944Abstract: In one embodiment, the present invention includes a method for training a Support Vector Machine (SVM) on a subset of features (d?) of a feature set having (d) features of a plurality of training instances to obtain a weight per instance, approximating a quality for the d features of the feature set using the weight per instance, ranking the d features of the feature set based on the approximated quality, and selecting a subset (q) of the features of the feature set based on the ranked approximated quality. Other embodiments are described and claimed.Type: ApplicationFiled: December 22, 2011Publication date: April 19, 2012Inventors: Eyal Krupka, Aharon Bar-Hillel
-
Patent number: 8108324Abstract: In one embodiment, the present invention includes a method for training a Support Vector Machine (SVM) on a subset of features (d?) of a feature set having (d) features of a plurality of training instances to obtain a weight per instance, approximating a quality for the d features of the feature set using the weight per instance, ranking the d features of the feature set based on the approximated quality, and selecting a subset (q) of the features of the feature set based on the ranked approximated quality. Other embodiments are described and claimed.Type: GrantFiled: May 15, 2008Date of Patent: January 31, 2012Assignee: Intel CorporationInventors: Eyal Krupka, Aharon Bar-Hillel
-
Publication number: 20090287621Abstract: In one embodiment, the present invention includes a method for training a Support Vector Machine (SVM) on a subset of features (d?) of a feature set having (d) features of a plurality of training instances to obtain a weight per instance, approximating a quality for the d features of the feature set using the weight per instance, ranking the d features of the feature set based on the approximated quality, and selecting a subset (q) of the features of the feature set based on the ranked approximated quality. Other embodiments are described and claimed.Type: ApplicationFiled: May 15, 2008Publication date: November 19, 2009Inventors: Eyal Krupka, Aharon Bar-Hillel