公开(公告)号：US20180330194A1

公开(公告)日：2018-11-15

申请号：US15719762

申请日：2017-09-29

Applicant: Siemens Aktiengesellschaft

Inventor： Kuan-Chuan Peng ,  Ziyan Wu ,  Jan Ernst

IPC: G06K9/62 ,  G06N3/08 ,  G06N3/04 ,  G06K9/00

CPC classification number: G06K9/6256 ,  G06K9/00201 ,  G06K9/4628 ,  G06K9/6267 ,  G06K9/6288 ,  G06N3/04 ,  G06N3/08

Abstract: Embodiments of the present invention provide a computer-implemented method for training an RGB-D classifier for a scene classification task. The method receives task-relevant labeled depth data, task-irrelevant RGB-D data, and a given trained representation in RGB. The method simulates an RGB representation using only the task-irrelevant RGB-D data. The method builds a joint neural network using only the task-irrelevant RGB-D data and the task-relevant labeled depth data.

公开(公告)号：US20180329041A1

公开(公告)日：2018-11-15

申请号：US15968913

申请日：2018-05-02

Applicant: TOPCON Corporation

Inventor： Fumio Ohtomo ,  Kaoru Kumagai ,  Nobuyuki Nishita

IPC: G01S7/481 ,  G01S17/42

CPC classification number: G01C15/002 ,  G01C15/06 ,  G06K9/00201

Abstract: A target instrument has an object to be measured which is provided with a reflection sheet, wherein the total station has a distance measuring light projecting unit, a light receiving unit, a distance measuring unit for performing a distance measurement of an object to be measured, an optical axis deflector capable of deflecting the distance measuring optical axis two-dimensionally, a projecting direction detecting module for detecting a deflection angle of the distance measuring optical axis and performing an angle measurement, and an arithmetic control module for controlling the optical axis deflector and the distance measuring unit, wherein the arithmetic control module is configured to two-dimensionally scan the object to be measured with the distance measuring light and to perform a distance measurement and an angle measurement with respect to the object to be measured, and further to detect a tilt and a tilt direction of the target instrument.

公开(公告)号：US20180326591A1

公开(公告)日：2018-11-15

申请号：US15974482

申请日：2018-05-08

Applicant: Atensor Engineering And Technology Systems GmbH

Inventor： Kurt Häusler

IPC: B25J11/00 ,  B25J9/16

CPC classification number: B25J11/0065 ,  B05D5/005 ,  B24B51/00 ,  B25J9/1664 ,  B25J9/1697 ,  B62D65/00 ,  G01B11/24 ,  G01B11/25 ,  G01C11/00 ,  G01N21/8806 ,  G01N21/95 ,  G05B19/4097 ,  G05B2219/35012 ,  G05B2219/45058 ,  G06K9/00201 ,  G06K9/03

Abstract: A method for automated detection of defects in a workpiece surface and generation of a robot program for the machining of the workpiece is described. In accordance with one embodiment, the method comprises the localization of defects in a surface of a workpiece as well as determining a three-dimensional topography of the localized defects and categorizing at least one localized defect based on its topography. Dependent on the defect category of the at least one defect, a machining process is selected and, in accordance with the selected machining process, a robot program for the robot-assisted machining of the at least one defect is generated with the assistance of a computer.

公开(公告)号：US20180315221A1

公开(公告)日：2018-11-01

申请号：US15582987

申请日：2017-05-01

Applicant: Lockheed Martin Corporation

Inventor： Michael Jones ,  Adam James Dickin

IPC: G06T11/00 ,  G06K9/62 ,  G06T7/73 ,  G06F17/30

CPC classification number: G06T11/005 ,  G06F17/30256 ,  G06F17/30268 ,  G06F17/3028 ,  G06K9/00201 ,  G06K9/0063 ,  G06K9/6201 ,  G06K9/6211 ,  G06K9/6807 ,  G06T7/74 ,  G06T15/005 ,  G06T15/205 ,  G06T17/20 ,  G06T2207/10028 ,  G06T2207/30244 ,  G06T2210/52 ,  G06T2215/16

Abstract: A system provides camera position and point cloud estimation 3D reconstruction. The system receives images and attempts existing structure integration to integrate the images into an existing reconstruction under a sequential image reception assumption. If existing structure integration fails, the system attempts dictionary overlap detection by accessing a dictionary database and searching to find a closest matching frame in the existing reconstruction. If overlaps are found, the system matches the images with the overlaps to determine a highest probability frame from the overlaps, and attempts existing structure integration again. If overlaps are not found or existing structure integration fails again, the system attempts bootstrapping based on the images. If any of existing structure integration, dictionary overlap detection, or bootstrapping succeeds, and if multiple disparate tracks have come to exist, the system attempts reconstructed track merging.

公开(公告)号：US20180292286A1

公开(公告)日：2018-10-11

申请号：US15800116

申请日：2017-11-01

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： MATTHIAS DITTBERNER ,  LEVENTE KLEIN ,  JASON D. RENWICK

IPC: G01M3/04 ,  G01N33/00 ,  G01C21/00

CPC classification number: G01N33/0047 ,  B64C39/024 ,  B64C2201/024 ,  B64C2201/123 ,  B64C2201/127 ,  B64C2201/141 ,  G01C21/00 ,  G01M3/04 ,  G01N21/3504 ,  G01N2021/0143 ,  G01N2021/1795 ,  G01N2201/0214 ,  G05D1/0088 ,  G05D1/0094 ,  G05D1/101 ,  G06K9/00201 ,  G06K9/0063 ,  G06K9/2018 ,  G06T7/73 ,  G06T17/05 ,  G06T2207/10032 ,  G06T2207/10048 ,  G06T2207/30181 ,  G08G5/0013 ,  G08G5/0021 ,  G08G5/0034 ,  G08G5/0039 ,  G08G5/0069 ,  G08G5/0086

Abstract: Methods, systems and computer program products for detecting gas leaks using a drone are provided. Aspects include capturing a first set of data regarding a presence of a gas in the geographic area while flying along the initial flight path. Aspects also include creating secondary flight paths through regions in the geographic area in which the presence of the gas exceeds a threshold amount and capturing a second set of data regarding a concentration of the gas in the one or more regions while flying along the secondary flight paths. Aspects further include capturing wind data while flying along the initial and second flight paths and creating a three-dimensional gas plume model for gas leaks identified in the geographic area based on the first set of data, the second set of data and the wind data, wherein the three-dimensional gas plume model identifies a source of the gas leaks.

公开(公告)号：US20180288431A1

公开(公告)日：2018-10-04

申请号：US15939098

申请日：2018-03-28

Applicant: NVIDIA Corporation

Inventor： Ming-Yu Liu ,  Xiaodong Yang ,  Jan Kautz ,  Sergey Tulyakov

IPC: H04N19/513 ,  G06K9/00 ,  G06N3/08 ,  G06T13/40

CPC classification number: H04N19/521 ,  G06K9/00201 ,  G06K9/00281 ,  G06N3/0445 ,  G06N3/0454 ,  G06N3/0472 ,  G06N3/08 ,  G06T13/40 ,  G06T2207/20081 ,  G06T2207/30196

Abstract: A method, computer readable medium, and system are disclosed for action video generation. The method includes the steps of generating, by a recurrent neural network, a sequence of motion vectors from a first set of random variables and receiving, by a generator neural network, the sequence of motion vectors and a content vector sample. The sequence of motion vectors and the content vector sample are sampled by the generator neural network to produce a video clip.

公开(公告)号：US20180268201A1

公开(公告)日：2018-09-20

申请号：US15888629

申请日：2018-02-05

Applicant: NEC Laboratories America, Inc.

Inventor： Xiang Yu ,  Kihyuk Sohn ,  Manmohan Chandraker

IPC: G06K9/00 ,  G06T19/20 ,  G06K9/46

CPC classification number: G06K9/00288 ,  G06F16/71 ,  G06F16/743 ,  G06F16/784 ,  G06K9/00201 ,  G06K9/00208 ,  G06K9/00214 ,  G06K9/00255 ,  G06K9/00275 ,  G06K9/00771 ,  G06K9/00899 ,  G06K9/4628 ,  G06K9/6256 ,  G06T19/20 ,  G06T2210/44

Abstract: A face recognition system is provided. The system includes a device configured to capture an input image of a subject. The system further includes a processor. The processor estimates, using a 3D Morphable Model (3DMM) conditioned Generative Adversarial Network, 3DMM coefficients for the subject of the input image. The subject varies from an ideal front pose. The processor produces, using an image generator, a synthetic frontal face image of the subject of the input image based on the input image and the 3DMM coefficients. An area spanning the frontal face of the subject is made larger in the synthetic image than in the input image. The processor provides, using a discriminator, a decision indicative of whether the subject of the synthetic image is an actual person. The processor provides, using a face recognition engine, an identity of the subject in the input image based on the synthetic and input images.

公开(公告)号：US20180268195A1

公开(公告)日：2018-09-20

申请号：US15980701

申请日：2018-05-15

Applicant: SHENZHEN UNIVERSITY

Inventor： Sen Jia ,  Jie Hu ,  Yao Xie ,  Linlin Shen

IPC: G06K9/00 ,  G06K9/62 ,  G06K9/46 ,  G06F17/18

CPC classification number: G06K9/0063 ,  G06F17/18 ,  G06K9/00201 ,  G06K9/46 ,  G06K9/4619 ,  G06K9/6228 ,  G06K9/6234 ,  G06K9/6267 ,  G06K2009/00644 ,  G06K2009/4657

Abstract: The present invention provides a Gabor cube feature selection-based classification method for hyperspectral remote sensing images, comprising the following steps: generating three-dimensional Gabor filters according to set frequency and direction parameter values; convoluting hyperspectral remote sensing images with the three-dimensional Gabor filters to obtain three-dimensional Gabor features; selecting three-dimensional Gabor features, classification contribution degrees to various classes of which meet preset requirements, from the three-dimensional Gabor features; and classifying the hyperspectral remote sensing images by a multi-task joint sparse representation-based classification means by using the selected three-dimensional Gabor features. The present invention is based on the three-dimensional Gabor features, and the used three-dimensional Gabor features contain rich local change information of a signal and are competent in feature characterizing. Using a Fisher discriminant criterion not only makes full use of high-level semantics hidden among the features, but also eliminates redundant information and reduces the classification time complexity.

公开(公告)号：US20180268184A1

公开(公告)日：2018-09-20

申请号：US15982238

申请日：2018-05-17

Applicant: Berkshire Grey, Inc.

Inventor： Thomas Wagner ,  Kevin Ahearn ,  Benjamin Cohen ,  Michael Dawson-Haggerty ,  Christopher Geyer ,  Thomas Koletschka ,  Kyle Maroney ,  Matthew T. Mason ,  Gene Temple Price ,  Joseph Romano ,  Daniel Smith ,  Siddhartha Srinivasa ,  Prasanna Velagapudi ,  Thomas Allen

IPC: G06K7/14 ,  B25J9/16 ,  G06Q10/08 ,  G06K19/06 ,  G06K9/00 ,  B25J19/02 ,  G05B19/12 ,  B65G1/137

CPC classification number: G06K7/1443 ,  B07C5/36 ,  B25J9/1694 ,  B25J9/1697 ,  B25J19/022 ,  B25J19/023 ,  B65G1/1373 ,  G05B19/124 ,  G05B19/4183 ,  G05B2219/14005 ,  G05B2219/36371 ,  G05B2219/45045 ,  G05B2219/45047 ,  G05B2219/45063 ,  G06K7/10693 ,  G06K7/1404 ,  G06K7/1447 ,  G06K9/00033 ,  G06K9/00201 ,  G06K19/06028 ,  G06K2209/19 ,  G06Q10/087 ,  Y02P90/083 ,  Y02P90/10

Abstract: A robotic system is disclosed that include an articulated arm and a first perception system for inspecting an object, as well as a plurality of additional perception systems, each of which is arranged to be directed toward a common area in which an object may be positioned by the robotic arm such that a plurality of views within the common area may be obtained by the plurality of additional perception systems.

公开(公告)号：US20180247545A1

公开(公告)日：2018-08-30

申请号：US15966560

申请日：2018-04-30

Applicant: Loveland Innovations, LLC

Inventor： Jim Loveland ,  Leif Larson ,  Dan Christiansen ,  Tad Christiansen ,  Cam Christiansen

IPC: G08G5/00 ,  G06K9/00 ,  G06K9/20 ,  B64C39/02 ,  G08G5/04 ,  G06Q40/08

CPC classification number: G08G5/0069 ,  B64C39/024 ,  B64C2201/123 ,  B64C2201/127 ,  B64C2201/141 ,  B64D47/08 ,  G01C11/02 ,  G05D1/0094 ,  G06K9/00201 ,  G06K9/0063 ,  G06K9/00637 ,  G06K9/00805 ,  G06K9/2063 ,  G06K9/2081 ,  G06K9/209 ,  G06Q40/08 ,  G08G5/045 ,  H04N5/23203 ,  H04N7/183 ,  H04N7/185

Abstract: An unmanned aerial vehicle (UAV) assessment and reporting system may conduct micro scans of a wide variety of property types. Scan data from any of a wide variety of sensor types may be compared with profile data using computer vision techniques to identify characteristics, defects, damage, construction materials, and the like. A hierarchal structure of the scan data may reduce the size of data sets used for property identification. For example, identifying a material type of a property may reduce or eliminate the need to compare scan data with specific data profiles for defects not related to the identified material type. Similarly, identifying a particular characteristic of a property may narrow down the data sets of possible material types. A rule set evaluator may evaluate matched profile data to determine adaptive actions to modify the navigation or scanning process of the UAV.