公开(公告)号：US11915466B2

公开(公告)日：2024-02-27

申请号：US17338328

申请日：2021-06-03

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Xipeng Yang ,  Xiao Tan ,  Hao Sun ,  Hongwu Zhang

IPC: G06T7/246 ,  G06V10/44 ,  G06T7/215 ,  G06T7/73 ,  G06F18/213 ,  G06V10/25 ,  G06V30/19 ,  G06V30/24 ,  G06V10/82

CPC classification number: G06V10/454 ,  G06F18/213 ,  G06T7/215 ,  G06T7/246 ,  G06T7/74 ,  G06V10/25 ,  G06V10/82 ,  G06V30/19173 ,  G06V30/2504 ,  G06T2207/20016 ,  G06T2207/20081 ,  G06V2201/07

Abstract: Embodiments of the present disclosure disclose a method and apparatus for determining a target anchor, a device and a storage medium. The method may include: extracting a plurality of feature maps of an original image using a feature extraction network; inputting the plurality of feature maps into a feature pyramid network to perform feature fusion, to obtain a plurality of fused feature maps; and using a region proposal network to implement operations as follows: determining an initial anchor of a network head using the fused feature map, based on a size of each fused feature map, and determining an offset parameter of the initial anchor, based on a ratio of the size of the fused feature map to the original image, and generating a plurality of candidate anchors in different directions, based on the offset parameter of the initial anchor.

公开(公告)号：US20240062557A1

公开(公告)日：2024-02-22

申请号：US18259381

申请日：2021-12-24

Applicant: ArcSoft Corporation Limited

Inventor： Zhaohang Zhang ,  Jin Wang

IPC: G06V20/59 ,  G06V40/16 ,  G06V10/25 ,  G06V10/82 ,  B60W40/08 ,  B60W50/14

CPC classification number: G06V20/593 ,  G06V40/161 ,  G06V10/25 ,  G06V10/82 ,  B60W40/08 ,  B60W50/14 ,  G06V2201/07 ,  B60W2040/0881 ,  B60W2420/42 ,  B60W2540/223

Abstract: A safety belt wearing detection method includes: acquiring a target image collected from a traffic device; determining a first image region and/or a second image region in the target image, the first image region being a region where a human torso of a target object in the target image is located, the second image region being a region where a face of the target object in the target image is located; determining a third image region according to the first image region and/or the second image region, the third image region being a safety belt wearing region in the target image; and determining whether wearing of a safety belt in the target image meets a preset wearing requirement according to image information in the third image region.

公开(公告)号：US20240062552A1

公开(公告)日：2024-02-22

申请号：US18270106

申请日：2022-02-21

Applicant: ARGO AI GMBH

Inventor： Roland MEERTENS ,  Sergei NIKIFOROV

IPC: G06V20/56 ,  G06V10/60 ,  G06T9/00

CPC classification number: G06V20/56 ,  G06V10/60 ,  G06T9/00 ,  G06V2201/07

Abstract: The invention is concerned with a method for generating encoded training data, wherein a processing unit performs the following steps for generating the encoded training data (11) and for storing them into a data storage during a test drive of a test vehicle that carries a camera that is generating raw image data. It is tested, if a given video encoder is suitable for encoding the raw image data for generating the encoded training data for a training of another, second artificial neural network, by performing a testing procedure.

公开(公告)号：US20240061092A1

公开(公告)日：2024-02-22

申请号：US18269384

申请日：2021-12-06

Applicant: KOITO MANUFACTURING CO., LTD.

Inventor： Takanori NAMBA ,  Masaaki SATO ,  Kiyotaka MOCHIZUKI

IPC: G01S7/497 ,  G01S17/931 ,  G01S17/86 ,  G06T7/70 ,  G06V20/56

CPC classification number: G01S7/497 ,  G01S17/931 ,  G01S17/86 ,  G06T7/70 ,  G06V20/56 ,  G01S2007/4977 ,  G06T2207/30252 ,  G06V2201/07

Abstract: A cleaner for cleaning a transmissive portion of a sensor including a light receiving unit that receives light from a detection target via the transmissive portion is not operated by a cleaner control unit within a predetermined time after the drive of the cleaner is complete.

公开(公告)号：US11908245B2

公开(公告)日：2024-02-20

申请号：US17931268

申请日：2022-09-12

Applicant: KEPLER VISION TECHNOLOGIES B.V.

Inventor： Henricus Meinardus Gerardus Stokman ,  Marc Jean Baptist Van Oldenborgh ,  Fares Alnajar

IPC: G06V40/20 ,  G06T7/70 ,  G06N20/00

CPC classification number: G06V40/28 ,  G06N20/00 ,  G06T7/70 ,  G06T2207/10016 ,  G06T2207/20076 ,  G06T2207/30196 ,  G06V2201/07

Abstract: A body language system for determining a body language message of a living being in a context comprising an artificial intelligence system, said AI system running a computer program that: retrieves an image of said living being showing body language; labels said living being in said image, resulting in a labeled living being; determines said context from said image using a trained machine learning model; determines a baseline body language of said labeled living being from said image using a trained machine learning model; adapts a trained machine learning model of said AI system using said baseline body language and said context; applies the adapted trained machine learning model of said AI system to the one image for categorizing said body language resulting in a category, and applying said category for determining said body language message.

公开(公告)号：US20240056687A1

公开(公告)日：2024-02-15

申请号：US18496325

申请日：2023-10-27

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Kyungdong YANG ,  Soonae KIM

IPC: H04N23/695 ,  G06T7/70 ,  G06V10/25 ,  H04N23/61 ,  G02B7/18

CPC classification number: H04N23/695 ,  G06T7/70 ,  G06V10/25 ,  H04N23/61 ,  G02B7/18 ,  G06V2201/07

Abstract: An electronic device disclosed herein includes: a display, a memory, a processor, and a camera, the camera includes a lens unit including a lens, a prism rotatable in a first direction or a second direction, a driving unit including at least one coil configured to move the prism, a magnetic body coupled to the prism, a sensor unit including a sensor configured to detect a first amount of rotation of the prism in the first direction or a second amount of rotation thereof in the second direction, and an image sensor configured to generate an electrical signal using light passing through the lens unit and the prism, wherein the processor is configured to: determine a first region of interest (ROI) for execution of a specified function using the camera, determine the first amount of rotation or the second amount of rotation using a sensing value of the sensor unit, and determine a second region of interest by correcting the first region of interest of the camera using the first amount of rotation or the second amount of rotation.

公开(公告)号：US20240054787A1

公开(公告)日：2024-02-15

申请号：US17766541

申请日：2019-12-16

Applicant: HANWHA TECHWIN CO., LTD.

Inventor： Myung-hwa SON ,  Changseup KIM ,  Minsuk SUNG

IPC: G06V20/52 ,  G06V20/40

CPC classification number: G06V20/52 ,  G06V20/44 ,  G06V2201/07

Abstract: A surveillance camera system according to an embodiment of the present disclosure comprises: multiple surveillance cameras for capturing images of different surveillance regions to surveil a protection object; an event management server which is connected to the surveillance cameras through a communication network and receives a first event signal or a second event signal from at least one of the surveillance cameras; and a manager terminal for receiving event information corresponding to the second event signal from the event management server when the second event signal is generated, wherein the first event signal corresponds to a signal generated when the protection object is detected by at least one of the multiple surveillance cameras, and the second event signal corresponds to a signal generated when the protection object is detected by none of the multiple surveillance cameras during a preconfigured reference time.

公开(公告)号：US20240054750A1

公开(公告)日：2024-02-15

申请号：US18447647

申请日：2023-08-10

Applicant: CASIO COMPUTER CO., LTD.

Inventor： Takuma ITO

IPC: G06V10/25 ,  G06V20/52 ,  G06T7/70

CPC classification number: G06V10/25 ,  G06V20/52 ,  G06T7/70 ,  G06V2201/07

Abstract: An image processing device including a processing unit, in which in a position in a central part of a frame image region related to a plurality of frame images, an output target region smaller than the frame image region is set as a range for which image data is output, and an out-of-frame monitor region to be a target for detecting a subject is set in a position deviated in a predetermined direction from the central part, in a part of the frame image region outside the output target region.

公开(公告)号：US20240054614A1

公开(公告)日：2024-02-15

申请号：US18553640

申请日：2022-03-10

Applicant: BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY

Inventor： Douglas WILLIAMS ,  Ian KEGEL ,  Brahim ALLAN ,  Martin TRIMBY ,  Luke PILGRIM

IPC: G06T5/50 ,  G06T3/40 ,  G06T7/11 ,  G06V10/44

CPC classification number: G06T5/50 ,  G06T3/40 ,  G06T7/11 ,  G06V10/44 ,  G06V2201/07 ,  G06T2207/20221

Abstract: Methods and systems for deciding whether, when, for how long, and/or where insertable image content will be displayed on top of a video presentation (i.e. existing image content) are disclosed. This can be for the imminent placement of insertable image content or a delayed placement of insertable image content. The decision making process depends upon the generation of a ‘cool map’. A cool map is a weighted map of the screen real estate that shows the areas that it would be practical to place insertable image content.

公开(公告)号：US20240046648A1

公开(公告)日：2024-02-08

申请号：US17753641

申请日：2020-09-07

Applicant: SONY GROUP CORPORATION

Inventor： XI CHEN

IPC: G06V20/52 ,  G05D1/12

CPC classification number: G06V20/52 ,  G05D1/12 ,  G06V2201/07

Abstract: The present disclosure relates to a crewless aircraft, a control method, and a program that enable safe interaction with a recognition target.
On the basis of at least one of a camera image shot of a recognition target by a camera or a drive signal of a motor for flying, a state estimation unit estimates a state of the recognition target, and an action determination unit determines an action of an own aircraft according to the state of the recognition target estimated. The technique according to the present disclosure can be applied to, for example, a drone.