Patents Examined by Hadi Akhavannik
  • Patent number: 10885379
    Abstract: This disclosure relates to improved techniques for performing multi-view image clustering. The techniques described herein utilize machine learning functions to optimize the image clustering process. Multi-view features are extracted from a collection of images. A machine learning function is configured to jointly learn a fused binary representation that combines the multi-view features and one or more binary cluster structures that can be used to partition the images. A clustering function utilizes the fused binary representation and the one or more binary cluster structures to generate one or more image clusters based on the collection of images.
    Type: Grant
    Filed: September 4, 2018
    Date of Patent: January 5, 2021
    Inventors: Zheng Zhang, Li Liu, Jie Qin, Fan Zhu, Fumin Shen, Yong Xu, Ling Shao, Heng Tao Shen
  • Patent number: 10880492
    Abstract: Provided are a high dynamic range (HDR) image capturing device, which is capable of capturing an HDR image optimized in various situations, and a control method therefor.
    Type: Grant
    Filed: March 8, 2017
    Date of Patent: December 29, 2020
    Inventors: Kyungyeon Min, Wonkyun Kim
  • Patent number: 10872117
    Abstract: Systems and methods are provided for distributed video storage and search over edge computing devices having a short-term memory and a long-term memory. The method may comprise caching a first portion of data on a first device. The method may further comprise determining, at a second device, whether the first device has the first portion of data. The determining may be based on whether the first piece of data satisfies a specified criterion. The method may further comprise sending the data, or a portion of the data, and/or a representation of the data from the first device to a third device.
    Type: Grant
    Filed: August 6, 2018
    Date of Patent: December 22, 2020
    Assignee: NETRADYNE, INC.
    Inventors: Avneesh Agrawal, Arun Valiaparambil, Tejeswara Rao Gudena, Anirudh Maringanti, David Jonathan Julian
  • Patent number: 10872242
    Abstract: In order to reduce a load of checking a result of image analysis, an information processing apparatus detects an object in an image, acquires a position in a monitoring region, acquires a similarity between a target and the detected object, and changes a display format of the object displayed at a position on a region image based on a notice level.
    Type: Grant
    Filed: February 28, 2018
    Date of Patent: December 22, 2020
    Assignee: Canon Kabushiki Kaisha
    Inventor: Shinji Yamamoto
  • Patent number: 10867185
    Abstract: A system and method for identifying media segments using audio augmented image cross-comparison is disclosed, in which a media segment identifying system analyses both audio and video content, producing a unique identifier to compare with previously identified media segments in a media segment database. The characteristic landmark-linked-image-comparisons are constructed by first identifying pairs of audio landmarks separated by a characteristic, or landmark, time. Digital images associated the audio landmarks are then compared, with the combination providing a characteristic landmark-linked-image-comparison. The audio landmarks are audio peaks that exceed predetermined thresholds. A landmark-time is the time between adjacent pairs of audio peaks. The pair of images associated with the audio peaks are reduced in pixel size and converted to gray scale. Corresponding pixels are compared to form a numeric comparison. One image may be mirrored before comparison to reduce the possibility of null comparisons.
    Type: Grant
    Filed: April 21, 2019
    Date of Patent: December 15, 2020
    Inventors: Samuel Chenillo, Oren Steinfeld
  • Patent number: 10867443
    Abstract: The present invention relates to an illustration of an annotation (22) at a sample slide (14). It is the intention for the invention to copy the annotation (22) provided at a reference slide (12) to the sample slide (14). Typically, a reference image (26) of the reference slide (12) is provided, in particular by scanning the reference slide (12). The reference image (26) therefore has the information about a reference slice (16) which is carried by the reference slide (12). Furthermore, the reference slide (12) has been marked with a region of interest (20) and the annotation (22), which can be associated with the region of interest (20). If such reference slide (12) is scanned, the respective reference image (26) comprises the information about the marked region of interest (20) as well as the annotation (22), too. Furthermore, the sample slide (14) carrying a sample slice (18) can be scanned, in order to provide a sample image (28).
    Type: Grant
    Filed: July 12, 2016
    Date of Patent: December 15, 2020
    Inventors: Jelte Peter Vink, Ruud Vlutters, Anke Pierik, Bas Hulsken, Reinhold Wimberger-Friedl, Sieglinde Neerken
  • Patent number: 10867182
    Abstract: An object recognition method and system thereof are provided. A recognition result of a first object of a (i?1)th frame of a video stream is obtained. A ith frame is received, and a second object is detected from the ith frame. Whether the first object and the second object are corresponding to the same target object is determined according to a position of the first object in the (i?1)th frame and a position of the second object in the ith frame. If the first object and the second object are corresponding to the same target object, whether a recognition confidence level is greater than a predetermined threshold is determined so as to perform the object recognition on the second object or assign the recognition result of the first object to the second object.
    Type: Grant
    Filed: August 16, 2018
    Date of Patent: December 15, 2020
    Inventors: Keng-Chih Chen, Jou-Yun Pan
  • Patent number: 10861174
    Abstract: A sampling and weighting technique is presented. Given a 3D model that is composed out of n separated entities, a set of parameters is obtained for each entity. A weight is calculated for each entity, giving higher weight for entities corresponding to rarer parameters. Entities are assigned to components based on their corresponding parameters. Entities are sampled based on the weights or based on the components. A new 3D model is constructed from the sampled entities.
    Type: Grant
    Filed: November 26, 2017
    Date of Patent: December 8, 2020
    Inventors: Vadim Kosoy, Dani Daniel
  • Patent number: 10842366
    Abstract: An endoscope system capable of setting the optimal balance of light source wavelengths in accordance with a diagnosis purpose is provided. An endoscope system includes a diagnosis purpose acquisition unit, a plurality of light sources with different light emission wavelengths, a light quantity ratio storage unit, a light quantity ratio selection unit, and a light source control unit. The diagnosis purpose acquisition unit acquires a diagnosis purpose. The light quantity ratio storage unit stores correspondence between the diagnosis purpose and a plurality of light quantity ratios with different balances of respective emission light quantities of the plurality of light sources. The light quantity ratio selection unit refers to the light quantity ratio storage unit and selects the light quantity ratio that is used for the acquired diagnosis purpose. The light source control unit controls the plurality of light sources to emit illumination light with the selected light quantity ratio.
    Type: Grant
    Filed: April 15, 2019
    Date of Patent: November 24, 2020
    Assignee: FUJIFILM Corporation
    Inventor: Tatsuya Aoyama
  • Patent number: 10835135
    Abstract: A non-contact heartbeat rate measurement system includes an image sensor, a target region selecting module, a heartbeat signal calculating module, a spectrum analyzing module, a vibration detecting module, a heartbeat peak selecting module. When a signal quality indicator is greater than a threshold value, and a first peak with global highest signal intensity in a heartbeat spectrum is similar to a face vibration frequency, the heartbeat peak selecting module selects a second peak with local highest signal intensity from a part frequency band of the heartbeat spectrum as an output heartbeat frequency. A non-contact heartbeat rate measurement method and an apparatus are disclosed herein.
    Type: Grant
    Filed: April 24, 2019
    Date of Patent: November 17, 2020
    Inventors: Bing-Fei Wu, Meng-Liang Chung, Tsong-Yang Tsou, Yun-Wei Chu, Kuan-Hung Chen, Po-Wei Huang, Yin-Yin Yang
  • Patent number: 10832030
    Abstract: A method of selecting a candidate fingerprint image for fingerprint recognition includes acquiring a target fingerprint image, generating a target frequency image based on a frequency conversion of the target fingerprint image, extracting frequency feature information associated with a representative frequency from the target frequency image, and selecting a candidate fingerprint image from enrolled fingerprint images by comparing the frequency feature information to enrolled frequency information.
    Type: Grant
    Filed: November 14, 2017
    Date of Patent: November 10, 2020
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Seon Min Rhee, Kyuhong Kim, Changkyu Choi
  • Patent number: 10832436
    Abstract: A method of recovering label positions in an imaging controller includes: receiving (i) a definition of a plane containing edges of the shelf, and (ii) a plurality of initial label indicators having locations on the plane; assigning the initial label indicators among a plurality of candidate subsets each representing a single physical label; for each candidate subset of initial indicators: generating, for each of a plurality of projection depths relative to the plane, a set of projections of the initial indicators in the candidate subset; determining an aggregate surface area for each set of projections; selecting, as a recovered depth for the candidate subset, one of the projection depths corresponding to the minimum aggregate surface area; and generating a recovered position for the candidate subset based on the recovered depth; and storing the recovered positions.
    Type: Grant
    Filed: April 5, 2018
    Date of Patent: November 10, 2020
    Assignee: Symbol Technologies, LLC
    Inventors: Joseph Lam, Vlad Gorodetsky, Richard Jeffrey Rzeszutek
  • Patent number: 10832055
    Abstract: Systems and methods for video presentation and analytics for live sporting events are disclosed. At least two cameras are used for tracking objects during a live sporting event and generate video feeds to a server processor. The server processor is operable to match the video feeds and create a 3D model of the world based on the video feeds from the at least two cameras. 2D graphics are created from different perspectives based on the 3D model. Statistical data and analytical data related to object movement are produced and displayed on the 2D graphics. The present invention also provides a standard file format for object movement in space over a timeline across multiple sports.
    Type: Grant
    Filed: January 30, 2019
    Date of Patent: November 10, 2020
    Inventor: Gerard J. Hall
  • Patent number: 10803542
    Abstract: One or more processors receives one or more first database records indicating a plurality of candidate identifiers are received by one or more processors. The one or more processors obtains one or more electronic images of the physical asset. At least one electronic image includes a graphical representation of data printed on the physical asset. The one or more processors identifies a plurality of regions of the data based on the one or more electronic images. The one or more processors determines a context of the physical asset based on the plurality of regions, selects the identifier from the plurality of candidate identifiers based on the plurality of regions, and generates a second database record including an indication of the context of the physical asset, and an indication of the identifier.
    Type: Grant
    Filed: September 14, 2018
    Date of Patent: October 13, 2020
    Assignee: LLC
    Inventors: Gerald Isaac Kestenbaum, Alexander D. Gaudio
  • Patent number: 10796159
    Abstract: In one aspect, a method includes receiving first query fingerprint data representing first content channeled through a portion of a content-distribution system. The method also includes detecting a first match between the received first query fingerprint data and first reference fingerprint data representing a modifiable content-segment. The method also includes responsive to detecting the first match, performing a first action. The method also includes receiving second query fingerprint data representing content received by a content-presentation device. The method also includes detecting a second match between the received second query fingerprint data and second reference fingerprint data representing second content transmitted by the content-distribution system, where the second content is a modified version of the first content. The method also includes responsive to detecting the second match, performing a second action that is different from the first action.
    Type: Grant
    Filed: March 31, 2020
    Date of Patent: October 6, 2020
    Assignee: The Nielsen Company (US), LLC
    Inventors: Andrew Laurence Ashbacher, Steven Michael Cormie, John L. Haynes, III, Stefan Matthijs Maris
  • Patent number: 10796179
    Abstract: The present application discloses a live human face verification method and device. The device acquires face images captured by at least two cameras and performs feature point registration on the face images according to preset face feature points, to obtain corresponding feature point combinations between the face images. After fitting out a homography transformation matrix among the feature point combinations, the device calculates transformation errors of the feature point combinations using the homography transformation matrix to obtain an error calculation result, and performing live human face verification of the face images according to the error calculation result. The embodiments of the present application do not need to calibrate the cameras, so the amount of calculation of a living body judgment algorithm can be reduced; moreover, the cameras can be freely placed, and thereby the flexibility and convenience of living body judgment can be increased.
    Type: Grant
    Filed: June 29, 2018
    Date of Patent: October 6, 2020
    Inventors: Guanbo Bao, Jilin Li
  • Patent number: 10796450
    Abstract: A method for detecting and tracking human head in an image by an electronic device is disclosed. The method may include segmenting the image into one or more sub-images; inputting each sub-image to a convolutional neural network trained according to training images having marked human head positions; outputting by a preprocessing layer of the convolutional neural network comprising a first convolutional layer and a pooling layer, a first feature corresponding to each sub-image; mapping through a second convolutional layer the first feature corresponding to each sub-image to a second feature corresponding to each sub-image; mapping through a regression layer the second feature corresponding to each sub-image to a human head position corresponding to each sub-image and a corresponding confidence level of the human head position; and filtering, according to the corresponding confidence level, human head positions corresponding to the one or more sub-images, to acquire detected human head positions in the image.
    Type: Grant
    Filed: March 12, 2019
    Date of Patent: October 6, 2020
    Assignee: Tencent Technology (Shenzhen) Company Limited
    Inventor: Deqiang Jiang
  • Patent number: 10796465
    Abstract: A method for establishing and/or reducing artifacts that arise when reconstructing (R) an MPI overall image data record from MPI signal data and an appended system matrix (SMapp(r,f)), wherein the appended system matrix (SMapp(r,f)) includes system matrices of different particle classes (P1, P2). The method includes setting a selection region (cvn(r)) of a reconstructed MPI overall image data record (cn(r)), producing virtual signal data (sn(f)) by inverse transformation of the selection region, reconstructing a virtual overall image data record (c?vn(r)) from the virtual signal data and the appended system matrix, setting an artifact region (cartifact_n(r)) within the reconstructed virtual overall image data record so that the artifact region has only voxels lying outside of the selection region, and assigning the image data present in the artifact region as artifact image data (cartifact_n(r)). This permits ghost artifacts to be corrected.
    Type: Grant
    Filed: March 13, 2019
    Date of Patent: October 6, 2020
    Inventors: Jochen Franke, Michael Herbst
  • Patent number: 10789739
    Abstract: The present disclosure relates to the generation of partial surface models them volumetric datasets for subsequent registration of such partial surface models to surface topology datasets. Specifically, given an object that is imaged using surface topology imaging and another volumetric modality, the volumetric dataset is processed in combination with an approach viewpoint to generate one or more partial surfaces of the object that will be visible to the surface topology imaging system. This procedure can eliminate internal structures from the surfaces generated from volumetric datasets, thus increases the similarity of the dataset between the two different modalities, enabling improved and quicker registration.
    Type: Grant
    Filed: July 12, 2019
    Date of Patent: September 29, 2020
    Assignee: 7D SURGICAL INC.
    Inventors: Adrian Linus Dinesh Mariampillai, Peter Siegler, Michael Leung, Beau Anthony Standish, Victor X. D. Yang
  • Patent number: 10789477
    Abstract: A method for real-time detection of at least one scene by an apparatus, from among a set of possible reference scenes, includes acquiring current values of attributes from measurement values supplied by sensors. The method further includes traversing a path through a decision tree. The nodes of the decision tree are respectively associated with the attributes. The traversal considers at each node along the path, the current value of the corresponding attribute, so as to obtain at the output of the path, a scene from among the set of reference scenes. The obtained scene identifying which reference scene is the detected scene. The method further includes developing a confidence index (SC) associated with the identification of the detected scene.
    Type: Grant
    Filed: March 27, 2018
    Date of Patent: September 29, 2020
    Inventors: Pierre Demaj, Laurent Folliot