Patents Examined by Jonathan S Lee
  • Patent number: 10733756
    Abstract: A computer-implemented method, system, and computer program product are provided for object detection utilizing an online flow guided memory network. The method includes receiving, by a processor, a plurality of videos, each of the plurality of videos including a plurality of frames. The method also includes generating, by the processor with a feature extraction network, a frame feature map for a current frame of the plurality of frames. The method additionally includes determining, by the processor, a memory feature map from the frame feature map and a previous memory feature map from a previous frame by warping the previous memory feature map. The method further includes predicting, by the processor with a task network, an object from the memory feature map. The method also includes controlling an operation of a processor-based machine to react in accordance with the object.
    Type: Grant
    Filed: August 29, 2018
    Date of Patent: August 4, 2020
    Assignee: NEC Corporation
    Inventors: Wongun Choi, Samuel Schulter, Tuan Hung Vu, Manmohan Chandraker
  • Patent number: 10733739
    Abstract: The present disclosure provides a method and system for displaying a target image based on a robot. The method includes successively acquiring video frame images; when a following target in an acquired kth-frame picture is detected, detecting information of a position of the following target in the kth-frame picture; displaying the kth-frame picture at a position of a k+N+1th frame in a time axis, and marking the position of the following target in the kth-frame picture, N being a number of pictures acquired within a detection cycle; sequentially predicting positions of the following target in acquired k+N+1th to k+2N?1th-frame pictures according to the position information of the following target in the kth-frame picture; and sequentially displaying the k+N+1th to k+2N?1th-frame pictures and the predicted positions of the following target at positions of the k+N+2th to k+2Nth frames in the time axis.
    Type: Grant
    Filed: May 4, 2018
    Date of Patent: August 4, 2020
    Assignee: Nanjing AvatarMind Robot Technology Co., Ltd.
    Inventors: Yangyang Xie, Changye Guo
  • Patent number: 10733484
    Abstract: An approach is provided for dynamic adaptation of an in-vehicle feature detector. The approach involves embedding a feature detection model, precomputed weights for the feature detection model, or a combination thereof in a data layer of map data representing a geographic area from which a training data set was collected to generate the feature detection model, the precomputed weights, or a combination thereof. The approach also involves deploying the feature detection model, the precomputed weights, or a combination thereof to adapt an in-vehicle feature detector based on determining that the in-vehicle feature detector is in the geographic area, plans to travel in the geographic area, or a combination thereof. The in-vehicle feature detector can then use the feature detection model, the precomputed weights, or a combination thereof to process sensor data collected while in the geographic area to detect one or more features.
    Type: Grant
    Filed: March 22, 2018
    Date of Patent: August 4, 2020
    Assignee: HERE Global B.V.
    Inventors: Vladimir Shestak, Stephen O'Hara, Nicholas Dronen
  • Patent number: 10726587
    Abstract: A method may include obtaining a first set of projection data with respect to a first dose level; reconstructing, based on the first set of projection data, a first image; determining a second set of projection data based on the first set of projection data, the second set of projection data relating to a second dose level that is lower than the first dose level; reconstructing a second image based on the second set of projection data; and training a first neural network model based on the first image and the second image. In some embodiments, the trained first neural network model may be configured to convert a third image to a fourth image, the fourth image exhibiting a lower noise level and corresponding to a higher dose level than the third image.
    Type: Grant
    Filed: October 25, 2017
    Date of Patent: July 28, 2020
    Assignee: SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD.
    Inventors: Qianlong Zhao, Guotao Quan, Xiang Li
  • Patent number: 10726568
    Abstract: A method for map constructing, applicable for real-time mapping of a to-be-localized area provided with at least one laser device, includes taking a position of a mobile electronic as a coordinate origin of a map coordinate system, when a center of a mark projected by a first laser device coincides with central point of CCD/CMOS; moving the mobile electronic device with the coordinate origin as a starting point to traverse the entire to-be-localized area, calculating and recording coordinate values of a position of one of obstacles each time when it is detected by the mobile electronic device; and constructing a map based on recorded information of mark and corresponding coordinate values and the coordinate values of the position of each said obstacle after the traversing process is finished.
    Type: Grant
    Filed: September 6, 2018
    Date of Patent: July 28, 2020
    Assignee: GUANGZHOU AIROB ROBOT TECHNOLOGY CO., LTD.
    Inventors: Yujie Jiang, Beichen Li
  • Patent number: 10719735
    Abstract: An information processing method, apparatus and video surveillance system have expandable functionality. The information processing method has an image acquisition step of acquiring an image; a first party feature quantity extracting step of extracting at least one first party feature quantity for describing a first party feature of the image from the acquired image; a third party feature quantity extracting step of invoking a third party algorithm to extract from the image at least one third party feature quantity for describing a third party feature other than the first party feature of the image; and a feature quantity correlation step of using the link information for describing the correlation between each of the first party feature quantities and each of the third party feature quantities of the same image to correlate each of the first party feature quantities of the image and each of the third party feature quantities.
    Type: Grant
    Filed: June 29, 2016
    Date of Patent: July 21, 2020
    Assignee: HITACHI, LTD.
    Inventors: Yulai Xie, Lu Wu, Peng Yang
  • Patent number: 10713763
    Abstract: In an embodiment, an image processing apparatus for performing geometrical transformation of data stored in accordance with two-dimensional coordinate positions includes: a control unit configured to decompose a transformation processing from first coordinates to second coordinates into linear transformation and non-linear transformation, the control unit also being configured to determine transformation parameters for the linear transformation and the non-linear transformation; a linear transformation unit configured to perform the linear transformation of the first coordinates to calculate third coordinates; a non-linear transformation unit configured to perform the non-linear transformation of the first coordinates to calculate fourth coordinates; and a combining unit configured to combine the third coordinates and the fourth coordinates to calculate the second coordinates.
    Type: Grant
    Filed: September 4, 2018
    Date of Patent: July 14, 2020
    Assignee: Canon Kabushiki Kaisha
    Inventor: Jumpei Ashida
  • Patent number: 10713471
    Abstract: A system and method for simulating facial expression of a virtual facial model are provided. The system stores a plurality of three-dimensional facial models corresponding to a plurality of preset sentiments one-to-one. The system identifies a present sentiment according to an acoustic signal and selects a selected model from the three-dimensional facial models according to the present sentiment, wherein the preset sentiment corresponding to the selected model is same as the present sentiment. The system predicts an upper half face image according to a lower half face image, combines the lower half face image and the upper half face image to form a whole face image, and generates a plurality of feature relationships by matching the facial features of the whole face image with the facial features of the selected model so that a virtual facial model can simulate an expression based on the feature relationships.
    Type: Grant
    Filed: November 20, 2018
    Date of Patent: July 14, 2020
    Assignee: Institute For Information Industry
    Inventors: Rong-Sheng Wang, Wan-Chi Ho, Hsiao-Chen Chang
  • Patent number: 10699433
    Abstract: A liquid level detecting method is provided. Firstly, a receiving component is provided, wherein the receiving component includes a transmissive container for containing a liquid and a marking component located on a rear surface of the transmissive container. Then, a camera captures a receiving component image of the receiving component from a front surface of the transmitting container, wherein the receiving component image includes a transmissive container image and a marking component image. Then, the marker component image is analyzed downward from an end portion to obtain multiple width values of multiple position points of the marker component image. Then, a width difference between each width value and a width average of the marking component image is obtained. Then, a position point whose width difference value is not within a preset range is determined. Then, according to the position point, a liquid level of the liquid is determined.
    Type: Grant
    Filed: November 23, 2018
    Date of Patent: June 30, 2020
    Assignee: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE
    Inventors: Jia-Hao Bai, Tzu-Hang Huang, Chun-Jen Chen, Cheng-Hsun Lin
  • Patent number: 10664996
    Abstract: In a method for the start-up operation of a multi-axis system, with the multi-axis system including, as components, segments connected via respective joints and are movable in one or more axes, and a tool, connected to one of the segments and is movable to a specified position, optical markers are arranged in the environment. Position coordinates of the optical markers in a first, global coordinate system are ascertained and stored in the controller. The environment is captured as image data by a camera system. The image data are transmitted to an AR system and visualized in an output apparatus. The optical markers and virtual markers are represented during the visualization of the image data, wherein a respective virtual marker is assigned to an optical marker. A check is performed as to whether an optical marker and the virtual marker overlay one another in the visualized image data.
    Type: Grant
    Filed: November 20, 2017
    Date of Patent: May 26, 2020
    Assignee: SIEMENS AKTIENGESELLSCHAFT
    Inventors: Mehdi Hamadou, Jan Richter, Maximilian Walter
  • Patent number: 10650612
    Abstract: A system and method for visualizing data obtained by performing a three-dimensional scan with penetrating radiation. Raw density arrays are formed from the scan, each raw density array being a three-dimensional array. A processed density array is formed by one or more operations, such as taking the difference between two raw density arrays, rotating the processed density array, multiplying the processed density array by a front-lighting array, and projecting the processed density array onto a plane to form an image, the projecting including calculating one or more of a plurality of statistics for each of a set of vectors each corresponding to a pixel of the image, the plurality of statistics including a vector mean, a vector maximum, and a vector standard deviation.
    Type: Grant
    Filed: June 7, 2019
    Date of Patent: May 12, 2020
    Assignee: Raytheon Company
    Inventors: Raymond Samaniego, John L. Tomich
  • Patent number: 10650280
    Abstract: In various embodiments, training objects are classified by human annotators, psychometric data characterizing the annotation of the training objects is acquired, a human-weighted loss function based at least in part on the classification data and the psychometric data is computationally derived, and one or more features of a query object are computationally classified based at least its part on the human-weighted loss function.
    Type: Grant
    Filed: April 11, 2019
    Date of Patent: May 12, 2020
    Assignee: PRESIDENT AND FELLOWS OF HARVARD COLLEGE
    Inventors: David Cox, Walter Scheirer, Samuel Anthony, Ken Nakayama
  • Patent number: 10650256
    Abstract: Among other things, one or more travel signals are identified by analyzing one or more images and data from sensors, classifying candidate travel signals into zero, one or more true and relevant travel signals, and estimating a signal state of the classified travel signals.
    Type: Grant
    Filed: April 18, 2017
    Date of Patent: May 12, 2020
    Assignee: nuTonomy Inc.
    Inventors: Baoxing Qin, Aravindkumar Vijayalingam
  • Patent number: 10643084
    Abstract: Among other things, one or more travel signals are identified by analyzing one or more images and data from sensors, classifying candidate travel signals into zero, one or more true and relevant travel signals, and estimating a signal state of the classified travel signals.
    Type: Grant
    Filed: April 18, 2017
    Date of Patent: May 5, 2020
    Assignee: nuTonomy Inc.
    Inventors: Baoxing Qin, Aravindkumar Vijayalingam
  • Patent number: 10635886
    Abstract: The present invention solves the problem by having: a first recording unit which accepts a handwritten electronic signature by a transactor and records the handwritten electronic signature in an electronic document; a second recording unit which records record information including transaction position information at a time that the handwritten electronic signature is recorded, in image data of the electronic document where the handwritten electronic signature is recorded by the first recording unit; and a managing unit which manages the record information recorded by the second recording unit and the image data where the record information is recorded as managing information, while associating the record information with the image data.
    Type: Grant
    Filed: September 15, 2015
    Date of Patent: April 28, 2020
    Assignee: NS SOLUTIONS CORPORATION
    Inventor: Shinichiro Tanaka
  • Patent number: 10636127
    Abstract: A method and system for bidirectional edge highlighting. A memory receives an original input image. A processor calculates edge intensity for each edge pixel in the original input image. The processor also identifies strong edges and attenuates said strong edges. The processor can further determine whether the attenuated strong edges should be bright or dark based on the intensity of the pixel in the original image. The processor applies bidirectional edge highlighting to the image based on the intensity of the pixel in the original image. The degree of darkness or brightness of the highlighting is based on the calculated edge intensity after strong edge attenuation.
    Type: Grant
    Filed: September 25, 2018
    Date of Patent: April 28, 2020
    Assignee: United States of America as represented by Secretary of the Navy
    Inventor: Gregory K. Fleizach
  • Patent number: 10628940
    Abstract: Systems and methods are provided for automated analysis of angiographic images. An angiographic imaging system is configured to capture a first image of a region of interest, representing a first time, and a second image of a region of interest, representing a second time. A registration component is configured to register the first image to the second image. A difference component is configured to generate a difference image from the first image and the second image. A pattern recognition component is configured to assign a clinical parameter to the region of interest from the difference image and at least one of the first image and the second image.
    Type: Grant
    Filed: September 8, 2015
    Date of Patent: April 21, 2020
    Assignee: THE CLEVELAND CLINIC FOUNDATION
    Inventors: Sunil K. Srivastava, Justis P. Ehlers
  • Patent number: 10630891
    Abstract: Provided is a an image processing apparatus including a candidate detection unit configured to detect each of candidate images serving as candidates for a main subject for a plurality of frames of image data, and a main subject determination unit configured to obtain a degree of stable presence of the candidate images detected by the candidate detection unit within the image data spanning the plurality of frames and to determine a main subject among the candidate images using the degree of stable presence.
    Type: Grant
    Filed: June 7, 2017
    Date of Patent: April 21, 2020
    Assignee: Sony Corporation
    Inventor: Masaya Kinoshita
  • Patent number: 10628916
    Abstract: An image generating apparatus for generating an output image based on an input panorama image, includes a parameter input unit inputting an output range parameter and a correction parameter, the output range parameter designating an output range in the panorama image, the correction parameter designating a correction part to be corrected in the output image; and an image correction unit correcting the correction part designated by the correction parameter in the output image. Further, the image correction unit calculates a similarity between the correction part and peripheral pixels and corrects the correction part based on the similarity and the peripheral pixels, and the output image is generated from the output range of the panorama image and is the panorama image corrected by the correction.
    Type: Grant
    Filed: November 8, 2018
    Date of Patent: April 21, 2020
    Assignees: Ricoh Company, Ltd., National University Corporation Nara Institute of Science and Technology
    Inventors: Daisuke Nakagawa, Norihiko Kawai, Tomokazu Sato, Naokazu Yokoya
  • Patent number: 10621746
    Abstract: Methods and apparatus for rapidly dimensioning an object are provided. An example method includes capturing, by a plurality of cameras, image data representative of an object; analyzing, by a processor, the image data to identify a plurality of candidate corners of the object; detecting, by the processor, a proximity of an appendage to each of the candidate corners; confirming, by the processor, based on respective proximities of the appendage to the candidate corners of the object, that a first one of the candidate corners is a corner of the object; and calculating, by the processor, based on the confirmed corner of the object, a dimension of the object.
    Type: Grant
    Filed: November 7, 2017
    Date of Patent: April 14, 2020
    Assignee: Symbol Technologies, LLC
    Inventors: Nicole Daphne Tricoukes, Michael J. Giannetta, Matthew Louis Kowalski