Patents Examined by Bernard Krasnic
  • Patent number: 10795979
    Abstract: In an approach to establishing personal identity using user identity patterns, one or more processors receive a set of input images corresponding to a period of time, where each input image in the set of input images corresponds to a specific time within the period of time. One or more processors may also identify a first user in the set of input images and determining, a user identity pattern based on the set of input images, where the user identity pattern includes multiple instances of at least one physical characteristic of the first user over the period of time. One or more processors may further determine a user behavior based on the user identity pattern. One or more processors may additionally associate the set of input images and the user identity pattern with a first user profile for the first user.
    Type: Grant
    Filed: September 27, 2017
    Date of Patent: October 6, 2020
    Assignee: International Business Machines Corporation
    Inventors: Yuk L. Chan, Deepti M. Naphade, Tin Hang To
  • Patent number: 10776663
    Abstract: A machine learning system can generate an image mask (e.g., a pixel mask) comprising pixel assignments for pixels. The pixels can be assigned to classes, including, for example, face, clothes, body skin, or hair. The machine learning system can be implemented using a convolutional neural network that is configured to execute efficiently on computing devices having limited resources, such as mobile phones. The pixel mask can be used to more accurately display video effects interacting with a user or subject depicted in the image.
    Type: Grant
    Filed: July 25, 2019
    Date of Patent: September 15, 2020
    Assignee: Snap Inc.
    Inventors: Lidiia Bogdanovych, William Brendel, Samuel Edward Hare, Fedir Poliakov, Guohui Wang, Xuehan Xiong, Jianchao Yang, Linjie Yang
  • Patent number: 10769800
    Abstract: A moving object detection apparatus has a generation unit that generates a background model based on a feature of a background region of a captured image that is captured by image capturing unit, a detection unit that detects a moving object region from an image input using an input unit, based on the background model, and a determination unit that determines whether to cause the generation unit to newly generate a background model, based on an amount of change in the moving object region detected by the detection unit for a first image and a second image captured at different times and input using the input unit.
    Type: Grant
    Filed: April 17, 2018
    Date of Patent: September 8, 2020
    Assignee: CANON KABUSHIKI KAISHA
    Inventor: Tomoya Honjo
  • Patent number: 10735820
    Abstract: An electronic device and a method for controlling the electronic device are provided. The method includes acquiring an image comprising a guide member positioned in an area at which a display device is to be installed; displaying the acquired image; analyzing a background area of the acquired image, the background area being positioned within the guide member in the acquired image; and acquiring information about a background image that is to be displayed in the background area of the acquired image.
    Type: Grant
    Filed: September 14, 2017
    Date of Patent: August 4, 2020
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Soo-hong Kim, Su-won Shin, Ki-suk Kim
  • Patent number: 10733434
    Abstract: A computer-implemented method, system and a computer program product are provided for automatically detecting redaction blocks in an image file document by analyzing the document to identify any redaction block areas and then detecting location information for each redaction block area identified in the document which may be mapped to any associated text fragments in the document based on the location information for each redaction block area and text fragment in the document.
    Type: Grant
    Filed: September 24, 2018
    Date of Patent: August 4, 2020
    Assignee: International Business Machines Corporation
    Inventors: Keith D. Cramer, Christophe Resse
  • Patent number: 10719958
    Abstract: A method relating to an image fusion includes acquiring a thermal infrared image and a visible image. The method also includes receiving a fusion parameter corresponding to a color space and generating, based on the fusion parameter, a fused image of the thermal infrared image and the visible image. The method further includes receiving a regulation parameter, the regulation parameter including a color scheme or a partial contrast, and adjusting the fused image according to the regulation parameter.
    Type: Grant
    Filed: September 27, 2016
    Date of Patent: July 21, 2020
    Assignee: ZHEJIANG DAHUA TECHNOLOGY CO., LTD.
    Inventors: Wei Lu, Qiankun Li
  • Patent number: 10713823
    Abstract: When a group of (pre-processed) projection data is stored into a projection-data storage unit, a Gaussian-based expansion-data creating unit creates a group of Gaussian-based expansion data that is expanded from each of the group of projection data through linear combination based on a plurality of Gaussian functions that is stored by a Gaussian-function storage unit and has different center points. A reconstruction-image creating unit then creates a reconstruction image by using the Gaussian-based expansion-data created by the Gaussian-based expansion-data creating unit, and stores the created reconstruction image into an image storage unit.
    Type: Grant
    Filed: June 2, 2017
    Date of Patent: July 14, 2020
    Assignee: TOSHIBA MEDICAL SYSTEMS CORPORATION
    Inventors: Manabu Teshigawara, Takuzo Takayama, Tomoyasu Komori, Takaya Umehara
  • Patent number: 10706593
    Abstract: The present invention provides a method for image reconstruction using target attribute assisted compressive sensing, including an initializing step, a subspace partitioning step, an atom set updating step, a sparse coefficient updating step and an outputting step. The present invention further provides a system for image reconstruction using target attribute assisted compressive sensing. A technical scheme provided by the present invention will introduce auxiliary information capable of reflecting target features into subspace partitioning in a case of unknown sparseness of a small target image signal, thereby accurately and effectively selecting the most closely matching dictionary subspace, and realizing efficient and rapid reconstruction of the small target image signal.
    Type: Grant
    Filed: September 4, 2018
    Date of Patent: July 7, 2020
    Assignee: SHENZHEN UNIVERSITY
    Inventors: Jianjun Huang, Runqing Liang, Li Kang, Zhongyin Liang
  • Patent number: 10699414
    Abstract: Image segmentation based on the combination of a deep learning network and a shape-guided deformable model is provided. In various embodiments, a time sequence of images is received. The sequence of images is provided to a convolutional network to obtain a sequence of preliminary segmentations. The sequence of preliminary segmentations labels a region of interest in each of the images of the sequence. A reference and auxiliary mask are generated from the sequence of preliminary segmentations. The reference mask corresponds to the region of interest. The auxiliary mask corresponds to areas outside the region of interest. A final segmentation corresponding to the region of interest is generated for each of the sequence of images by applying a deformable model to the composite mask with reference to the auxiliary mask.
    Type: Grant
    Filed: April 3, 2018
    Date of Patent: June 30, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Gopalkrishna Veni, Mehdi Moradi
  • Patent number: 10691984
    Abstract: There is disclosed a method of determining a digital document suitability for OCR processing, the method executable by a user electronic device, the user electronic device configured for capturing a digital image of a document. The method comprises: acquiring by the user electronic device, the digital image of the document; determining, by a classifier executed by the user electronic device, an OCR suitability parameter associated with the digital image, the OCR suitability parameter indicative of whether the digital image is suitable for producing an output of the OCR processing of an acceptable quality, the classifier having been trained to determine the OCR suitability parameter at least partially based on a level of noise associated with the digital image; in response to the OCR suitability parameter being below a pre-determined threshold, causing the user electronic device to re-acquire the digital image.
    Type: Grant
    Filed: December 13, 2016
    Date of Patent: June 23, 2020
    Assignee: ABBYY Production LLC
    Inventors: Vasily Loginov, Ivan Zagaynov
  • Patent number: 10692259
    Abstract: Techniques for automatic creation of media collages are described. In one or more implementations, unwanted frames are identified and removed from items of media content. A media score is then determined for items of media content based on characteristics of an appearance of the items within a plurality of collage templates. A template score is determined for each collage template of the plurality of collage templates by combining the media scores for each media item of the plurality of media items included in a collage template. At least one of the plurality of collage templates is selected based on determined template scores. Then, at least one media collage is outputted based on the selected collage templates.
    Type: Grant
    Filed: December 15, 2016
    Date of Patent: June 23, 2020
    Assignee: Adobe Inc.
    Inventors: Abhishek Shah, Sameer Bhatt
  • Patent number: 10679333
    Abstract: A defect in an image of a semiconductor wafer can be classified as an initial defect type based on the pixels in the image. Critical dimension uniformity parameters associated with the defect type can be retrieved from an electronic data storage unit. A level of defectivity of the defect can be quantified based on the critical dimension uniformity parameters. Defects also can be classified based on critical dimension attributes, topography attributes, or contrast attributes to determine a final defect type.
    Type: Grant
    Filed: July 26, 2018
    Date of Patent: June 9, 2020
    Assignee: KLA-Tencor Corporation
    Inventor: Hari Pathangi Sriraman
  • Patent number: 10679047
    Abstract: An object recognition system includes a parameter generator for generating a parameter that indicates whether an object depicted in an image pair is the same or different object, and a pose difference for the image pair and a parameter refiner for refining the parameter, and an applicator for applying the refined parameter to object recognition.
    Type: Grant
    Filed: December 29, 2017
    Date of Patent: June 9, 2020
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Rogerio Schmidt Feris, Minkyong Kim, Clifford A. Pickover
  • Patent number: 10671895
    Abstract: A “Best of Burst Selector,” or “BoB Selector,” automatically selects a subjectively best image from a single set of images of a scene captured in a burst or continuous capture mode, captured as a video sequence, or captured as multiple images of the scene over any arbitrary period of time and any arbitrary timing between images. This set of images is referred to as a burst set. Selection of the subjectively best image is achieved in real-time by applying a machine-learned model to the burst set. The machine-learned model of the BoB Selector is trained to select one or more subjectively best images from the burst set in a way that closely emulates human selection based on subjective subtleties of human preferences. Images automatically selected by the BoB Selector are presented to a user or saved for further processing.
    Type: Grant
    Filed: December 12, 2016
    Date of Patent: June 2, 2020
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Baoyuan Wang, Sing Bing Kang, Joshua Bryan Weisberg
  • Patent number: 10664978
    Abstract: A method for using a synthetically trained neural network for tracking particle movement in video microscopy data includes receiving, as input, video microscopy data representing images of particles that move between video frames. The method includes using a synthetically trained neural network to track movement of the particles between the video frames, wherein the synthetically trained neural network comprises a neural network that is trained on a plurality of different simulated video microscopy data sets. The method further includes outputting, by the synthetically trained neural network, an indication of movement of the particles between the video frames.
    Type: Grant
    Filed: April 9, 2019
    Date of Patent: May 26, 2020
    Assignee: The University of North Carolina at Chapel Hill
    Inventors: Jay Mack Newby, M. Gregory Forest, Samuel Kit Bunn Lai
  • Patent number: 10656270
    Abstract: This object detection device is provided with: an output unit; a plurality of detection units; a data acquisition unit; and an object information confirmation unit. The output unit outputs measurement light Lm towards an object. Each of the plurality of detection units detects, as a signal representing the distance to the object present in an observed area and the shape of the object, the reflected light generated as a result of the measurement light being reflected by the object. The data acquisition unit acquires detection signal waveforms W representing changes in the intensities of second signals over time. The object information confirmation unit confirms the range of the presence of the object by determining that any two or more of the detection units correspond to the same object, or correspond to different objects, on the basis of the detection signal waveforms.
    Type: Grant
    Filed: December 6, 2016
    Date of Patent: May 19, 2020
    Assignee: OMRON Corporation
    Inventor: Yukiko Yanagawa
  • Patent number: 10641888
    Abstract: Methods and systems for remote detection of objects involving cued sensor fusion. In some implementations, a first set of sensed data may be generated using a first object sensor and a second set of sensed data generated using a second object sensor. An object may then be detected using the sensed data from the first object sensor. Upon detecting the object using the sensed data from the first object sensor, a setting associated with the second object sensor may be changed to increase the probability of detecting the object using the second set of sensed data from the second object sensor.
    Type: Grant
    Filed: November 6, 2017
    Date of Patent: May 5, 2020
    Assignee: Veoneer US Inc.
    Inventors: Matthew Fetterman, Aret Carlsen
  • Patent number: 10643112
    Abstract: An online system distributes content items provided by content providers. The online system determines a likelihood of a content item having deceptive information. The online system stores images showing faces of people in an image database. The online system extracts features from the content item, and provides the extracted features to a machine learning based model configured to generate score indicating whether a content item comprises deceptive information. The machine learning based model uses at least a feature based on matching of faces of users shown in the content item with faces of users shown in the images of the image database. If the online system determines that a content item is deceptive, the online system adds images comprising faces extracted from the content item to the image database to grow the image database.
    Type: Grant
    Filed: March 27, 2018
    Date of Patent: May 5, 2020
    Assignee: Facebook, Inc.
    Inventors: Yang Mu, Daniel Olmedilla de la Calle
  • Patent number: 10621471
    Abstract: A system, method and computer program product is provided. An input signal for classification and a set of pre-classified signals are received, each comprising a vector representation of an object having a plurality of vector elements. A sparse vector comprising a plurality of sparse vector coefficients is determined. Each sparse vector coefficient corresponds to a signal in the set of pre-classified signals and represents the likelihood of a match between the object represented in the input signal and the object represented in the corresponding signal. A largest sparse vector coefficient is compared with a predetermined threshold. If the largest sparse vector coefficient is less than the predetermined threshold, the corresponding signal is removed from the set of pre-classified signals. The determining and comparing are repeated using the input signal and the reduced set of pre-classified signals.
    Type: Grant
    Filed: April 3, 2019
    Date of Patent: April 14, 2020
    Assignee: International Business Machines Corporation
    Inventors: Cecilia J. Aas, Raymond S. Glover
  • Patent number: 10621450
    Abstract: In a road shoulder extraction method, three-dimensional scan data of a plurality of frames is obtained. At least one high spatial point in the three-dimensional scan data of the plurality of frames is removed, to obtain three-dimensional scan data sets of the plurality of frames. A search is performed for a curvature change point and an elevation change point in each of the three-dimensional scan data sets of the plurality of frames. A search is performed for a road shoulder location point in the three-dimensional scan data set of each of the plurality of frames according to the curvature change point and the elevation change point in the three-dimensional scan data set of the respective frame. The road shoulder location point in the three-dimensional scan data set of each of the plurality of frames is extracted according to a preset algorithm. The extracted road shoulder location points are connected to obtain a road shoulder line.
    Type: Grant
    Filed: May 30, 2018
    Date of Patent: April 14, 2020
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventor: Chao Zeng