公开(公告)号：US20230230224A1

公开(公告)日：2023-07-20

申请号：US18149686

申请日：2023-01-04

Applicant: Siemens Aktiengesellschaft

Inventor： Baris Erol ,  Jason Dube ,  Yuan Zi

IPC: G06T7/00 ,  G06T7/33 ,  G06V10/25 ,  G06V10/74

CPC classification number: G06T7/0008 ,  G06T7/001 ,  G06T7/337 ,  G06V10/25 ,  G06V10/761 ,  G06V2201/07 ,  G06T2207/10024 ,  G06T2207/20081 ,  G06T2207/30136

Abstract: A method for detecting defects in a formed metal part includes locating one or more regions of interest in a synthetic image of a part manufactured by a forming process. The synthetic image is informed based on a physics-based simulation of the forming process. The regions of interest indicate a high risk of having a defect from the forming process. A set of training images including real images of actual manufactured parts are registered with the synthetic image. The regions of interest are overlaid on each training image, to extract patches from the training images that correspond to high-risk regions. An anomaly detection model is trained on the patches extracted from the training images to detect a defect in a formed metal part from an acquired image of the formed metal part, by detecting an anomaly in a patch extracted from the acquired image that corresponds to a high-risk region.

公开(公告)号：US20230229372A1

公开(公告)日：2023-07-20

申请号：US18102112

申请日：2023-01-27

Applicant: JVCKENWOOD Corporation

Inventor： Tetsuya Suwa ,  Hideo Tsuru ,  Motomu Takada ,  Takayuki Sugahara

IPC: G06F3/14 ,  G06F3/01 ,  G06V20/20 ,  G06Q30/0273

CPC classification number: G06F3/14 ,  G06F3/015 ,  G06V20/20 ,  G06Q30/0273 ,  G06V2201/07

Abstract: A display device includes: a biological sensor configured to detect biological information of a user; an output-specification determining unit configured to determine a display specification of a sub-image to be displayed on a display unit based on the biological information of the user; an output control unit configured to cause the display unit to display the sub-image in a superimposed manner on a main image that is visually recognized through the display unit and based on the display specification; and an environment sensor configured to detect environment information of a periphery of the display device. The environment information includes location information of the user, and the biological information includes brain wave information of the user. The output-specification determining unit is configured to determine display time of the sub-image per unit time as the display specification based on the location information and the brain wave information.

公开(公告)号：US11694388B2

公开(公告)日：2023-07-04

申请号：US17383465

申请日：2021-07-23

Applicant: Cognata Ltd.

Inventor： Dan Atsmon ,  Eran Asa ,  Ehud Spiegel

IPC: G06T15/10 ,  G06T7/55 ,  G06T7/579 ,  G06N3/08 ,  G06V20/56 ,  G06F18/21 ,  G06F18/24 ,  G06F18/28 ,  G05D1/00 ,  G05D1/02 ,  G06N3/04 ,  G06T3/00 ,  G06F18/214

CPC classification number: G06T15/10 ,  G05D1/0088 ,  G05D1/0246 ,  G06F18/217 ,  G06F18/2148 ,  G06F18/24 ,  G06F18/28 ,  G06N3/04 ,  G06N3/08 ,  G06T3/0018 ,  G06T7/55 ,  G06T7/579 ,  G06V20/56 ,  G05D2201/0213 ,  G06T2207/10028 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30252 ,  G06V2201/07

Abstract: A system for creating synthetic data for testing an autonomous system, comprising at least one hardware processor adapted to execute a code for: using a machine learning model to compute a plurality of depth maps based on a plurality of real signals captured simultaneously from a common physical scene, each of the plurality of real signals are captured by one of a plurality of sensors, each of the plurality of computed depth maps qualifies one of the plurality of real signals; applying a point of view transformation to the plurality of real signals and the plurality of depth maps, to produce synthetic data simulating a possible signal captured from the common physical scene by a target sensor in an identified position relative to the plurality of sensors; and providing the synthetic data to at least one testing engine to test an autonomous system comprising the target sensor.

公开(公告)号：US20230206485A1

公开(公告)日：2023-06-29

申请号：US17927216

申请日：2020-05-28

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Tianmin Rao

IPC: G06T7/73 ,  G06T7/62 ,  G06V10/25 ,  G06V10/44

CPC classification number: G06T7/73 ,  G06T7/62 ,  G06V10/25 ,  G06V10/44 ,  G06V2201/07 ,  G06T2207/30242

Abstract: Disclosed are a target detection method based on a heterogeneous platform, a terminal device and a storage medium. A specific implementation solution is as follows: a first processor receives a video stream image, and is able to first determine a number of targets in an Nth-frame image when performing target detection on an (N+K)th-frame image in the video stream image;
when the number of targets in the Nth-frame image is greater than or equal to a target threshold, the first processor is able to send the pre-processed (N+K)th-frame image to a second processor, so that the second processor cooperates with the first processor to complete target detection on the (N+K)th-frame image; and when the number of targets in the Nth-frame image is less than the target threshold, the first processor is able to perform target detection on the (N+K)th-frame image individually.

公开(公告)号：US20230206483A1

公开(公告)日：2023-06-29

申请号：US18088017

申请日：2022-12-23

Applicant: Rakuten Group, Inc.

Inventor： Mitsuru NAKAZAWA ,  Takashi TOMOOKA

IPC: G06T7/70 ,  G06T17/00 ,  G06T5/50 ,  G06T7/50 ,  G06V10/44

CPC classification number: G06T7/70 ,  G06T17/00 ,  G06T5/50 ,  G06T7/50 ,  G06V10/44 ,  G06V2201/07 ,  G06T2207/20221

Abstract: Disclosed is an image processing apparatus including a receiving section that receives a captured image of a target object, an instruction section that gives a user an instruction on which direction to capture an image of the target object, an inferencing section that performs an inference of an imaging direction on the basis of the received captured image, and a verifying section that performs predetermined verification processing related to the target object on the basis of an imaging direction as the instructed direction and the imaging direction as a result of the inference.

公开(公告)号：US20230206463A1

公开(公告)日：2023-06-29

申请号：US18088433

申请日：2022-12-23

Applicant: NUCTECH COMPANY LIMITED ,  Tsinghua University ,  Beijing Shenmutek Company Limited

Inventor： Ziran ZHAO ,  Jianping GU ,  Yingxin WANG ,  Shaoming YUAN ,  Li GU ,  Yitao JIAO

IPC: G06T7/20 ,  G06V10/44 ,  G06V10/74 ,  G01N23/04

CPC classification number: G06T7/20 ,  G06V10/44 ,  G06V10/761 ,  G01N23/04 ,  G06V2201/07 ,  G06T2207/30112 ,  G06T2207/30196

Abstract: Provided are a method and a device of binding security inspection information, an electronic apparatus and a storage medium, which may be applied to the field of security inspection technology. In the method of binding security inspection information of the present disclosure, the pedestrian detection frame is acquired by detecting the video stream, and each pedestrian detection frame has the coordinate information and the tracking information. According to the coordinate information and the tracking information of the pedestrian detection frame, the baggage security inspection information, the pedestrian security inspection information, and the identity information are bound to acquire all security inspection information of a detected person, so that a security inspection efficiency is improved.

公开(公告)号：US11686848B2

公开(公告)日：2023-06-27

申请号：US17007651

申请日：2020-08-31

Applicant: UATC, LLC

Inventor： Xuanyuan Tu ,  Sivabalan Manivasagam ,  Mengye Ren ,  Ming Liang ,  Bin Yang ,  Raquel Urtasun

IPC: G01S17/931 ,  G06N20/00 ,  G01S17/89 ,  G01S17/42 ,  G01S17/894 ,  G06V20/56 ,  G06V20/64 ,  G06F18/214 ,  G06F18/21 ,  G06V10/764 ,  G06V10/82

CPC classification number: G01S17/89 ,  G01S17/42 ,  G01S17/894 ,  G01S17/931 ,  G06F18/214 ,  G06F18/217 ,  G06N20/00 ,  G06V10/764 ,  G06V10/82 ,  G06V20/56 ,  G06V20/64 ,  G06V2201/07 ,  G06V2201/08

Abstract: Systems and methods for training object detection models using adversarial examples are provided. A method includes obtaining a training scene and identifying a target object within the training scene. The method includes obtaining an adversarial object and generating a modified training scene based on the adversarial object, the target object, and the training scene. The modified training scene includes the training scene modified to include the adversarial object placed on the target object. The modified training scene is input to a machine-learned model configured to detect the training object. A detection score is determined based on whether the training object is detected, and the machine-learned model and the parameters of the adversarial object are trained based on the detection output. The machine-learned model is trained to maximize the detection output. The parameters of the adversarial object are trained to minimize the detection output.

公开(公告)号：US20230196895A1

公开(公告)日：2023-06-22

申请号：US18085043

申请日：2022-12-20

Applicant: POSCO Co., Ltd ,  RESEARCH INSTITUTE OF INDUSTRIAL SCIENCE & TECHNOLOGY ,  POSTECH Research and Business Development Foundation

Inventor： jinwoo Lee ,  Daijin Kim ,  Dongmin Cha ,  Joonyeon Lee ,  Geon Tae Ahn ,  Hyeongjin Jeon

IPC: G08B21/02 ,  G08B7/06 ,  G06V20/52 ,  G06V40/10 ,  G06V10/82 ,  G06T7/20

CPC classification number: G08B21/02 ,  G08B7/06 ,  G06V20/52 ,  G06V40/10 ,  G06V10/82 ,  G06T7/20 ,  G06V2201/07 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30196 ,  G06T2207/30232

Abstract: A method for monitoring a state of wearing safety protective equipment according to an embodiment includes receiving an image from a site monitoring device installed at an operation site determining whether a work performed in a monitoring area corresponds to a dangerous work by analyzing the image; detecting a target worker and a body area of the target worker when it is determined that the work corresponds to the dangerous work as a result of determining whether the work performed in the monitoring area corresponds to the dangerous work; determining a type of safety protective equipment required for the target worker; and determining the state of the target worker wearing the safety protective equipment.

公开(公告)号：US20230196779A1

公开(公告)日：2023-06-22

申请号：US18066531

申请日：2022-12-15

Applicant: Commissariat à I'énergie atomique et aux énergies alternatives

Inventor： Maxence Bouvier ,  Alexandre Valentian

IPC: G06V20/52 ,  G06T7/20 ,  G06T7/579 ,  G06V10/60 ,  H04N25/772

CPC classification number: G06V20/52 ,  G06T7/20 ,  G06T7/579 ,  G06V10/60 ,  H04N25/772 ,  G06V2201/07

Abstract: The present invention relates to an observation system (10) of an environment, the observation system (10) comprising:



a sensor (12) forming a synchronous stream of framed data,
a first processing chain (14) including:
a first reception unit (30) receiving the formed synchronous stream, and
a first processing unit (32) comprising a first conversion block (34) converting the synchronous stream into event data, and a first calculation block (36) calculating first data from the event data
a second processing chain (16) including a second processing unit (46) receiving a synchronous stream of framed data from the array (18) and the first data and obtaining second data relating to the environment as a function of the synchronous stream and the first data.

公开(公告)号：US20230196731A1

公开(公告)日：2023-06-22

申请号：US17556662

申请日：2021-12-20

Applicant: GM CRUISE HOLDINGS LLC

Inventor： Pranay Agrawal ,  Cheng-An Hou ,  Debanjan Nandi

IPC: G06V10/77 ,  G06V10/80 ,  G06T7/73

CPC classification number: G06V10/7715 ,  G06V10/803 ,  G06T7/73 ,  G06T2210/12 ,  G06T2207/30252 ,  G06T2207/10028 ,  G06V2201/07

Abstract: The disclosed technology provides solutions for object detection and classification with both high recall and high precision, by using a first stage with high recall, and a second stage to provide high precision. The dimensional state of a pointcloud is reduced from 3 to 2, and proposed bounding boxes are generated. The original pointcloud data is filtered according to the bounding boxes, and fused with learned features, with the fused data processed to generate the high recall and high precision output.