知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

1259 records (took 0.02 seconds)

    IMAGE PROCESSING APPARATUS, IMAGE PROCESSING SYSTEM, IMAGE PROCESSING METHOD, AND NON-TRANSITORY RECORDING MEDIUM
    353.
    发明公开
    IMAGE PROCESSING APPARATUS, IMAGE PROCESSING SYSTEM, IMAGE PROCESSING METHOD, AND NON-TRANSITORY RECORDING MEDIUM 审中-公开

    公开(公告)号：US20240112353A1

    公开(公告)日：2024-04-04

    申请号：US18477953

    申请日：2023-09-29

    Applicant: Mei Oyama

    Inventor： Mei Oyama

    IPC: G06T7/20 ,  G06T7/70 ,  G06V10/25

    CPC classification number: G06T7/20 ,  G06T7/70 ,  G06V10/25 ,  G06T2207/30241 ,  G06V2201/07

    Abstract: An image processing apparatus communicates with an image capturing apparatus. The image processing apparatus includes circuitry to acquire an image of an imaging range of the image capturing apparatus, captured by the image capturing apparatus, recognize an identification that identifies an individual target object included in the image, calculate a trajectory of positions between which the target object included in the image moves, estimate an area in which the target object is present based on the trajectory, acquire the trajectory based on the positions at which the identification of the target object is recognized, and obtain individual area estimation information associating the estimated area corresponding to the acquired trajectory and the identification of the target object.

    Technique For Assigning Marker Identities To Markers Of A Tracker
    356.
    发明公开
    Technique For Assigning Marker Identities To Markers Of A Tracker 审中-公开

    公开(公告)号：US20240104747A1

    公开(公告)日：2024-03-28

    申请号：US18372905

    申请日：2023-09-26

    Applicant: Stryker European Operations Limited

    Inventor： Emeric Umbdenstock ,  Florian Franitza ,  Marvin Koepff

    IPC: G06T7/20 ,  G06T3/40 ,  G06V10/24 ,  H04N23/11

    CPC classification number: G06T7/20 ,  G06T3/40 ,  G06V10/242 ,  H04N23/11 ,  G06T2207/10048 ,  G06T2207/30204 ,  G06V2201/07

    Abstract: A method, computer program product, device, and tracking system for assigning marker identities to markers of a tracker are provided. The tracker includes a reference detectable in a visible light spectrum, and the markers are detectable at least in an infrared light spectrum. The markers are arranged in a pre-determined relationship relative to the reference and the pre-determined relationship is indicative of the marker identities. A method implementation includes receiving first image data of the markers captured in the infrared spectrum. The method further includes receiving second image data of the reference captured in the visible light spectrum. The first and second image data may be captured under at least essentially the same viewing angle. The method further includes assigning the marker identities to the markers determined in the first image data based on the reference determined in the second image data and the pre-determined relationship.

    SYSTEM AND METHOD FOR CAPTURING SOUND SOURCE
    358.
    发明公开
    SYSTEM AND METHOD FOR CAPTURING SOUND SOURCE 审中-公开

    公开(公告)号：US20240098406A1

    公开(公告)日：2024-03-21

    申请号：US18368799

    申请日：2023-09-15

    Applicant: AVER INFORMATION INC.

    Inventor： Pang-Ti TAI ,  Po-Hsun WU ,  Yun-Long SIE

    IPC: H04R1/32 ,  G06V10/25 ,  H04R1/08

    CPC classification number: H04R1/326 ,  G06V10/25 ,  H04R1/08 ,  G06V2201/07 ,  G06V2201/09 ,  H04R2201/40

    Abstract: A method for capturing a sound source includes: capturing a space where a microphone array is located to generate an image by a camera, wherein the microphone array is configured to receive a sound generated by the sound source and generate a sound source coordinate of the sound source relative to the microphone array; searching for a sub-image belonging to the microphone array within the images by a computing device connected to the camera; calculating a microphone coordinate of the microphone array relative to the camera by the computing device according to the sub-image; calculating a required control parameter by the computing device at least according to the sound source coordinate and the microphone coordinate; adjusting a capturing direction by the camera to capture the sound source at least according to the required control parameter.

    LANDMARK DETECTION WITH AN ITERATIVE NEURAL NETWORK
    359.
    发明公开
    LANDMARK DETECTION WITH AN ITERATIVE NEURAL NETWORK 审中-公开

    公开(公告)号：US20240096115A1

    公开(公告)日：2024-03-21

    申请号：US18243555

    申请日：2023-09-07

    Applicant: NVIDIA Corporation

    Inventor： Pavlo Molchanov ,  Jan Kautz ,  Arash Vahdat ,  Hongxu Yin ,  Paul Micaelli

    IPC: G06V20/59 ,  G06T7/70 ,  G06V10/82 ,  G06V20/70 ,  G06V40/16

    CPC classification number: G06V20/597 ,  G06T7/70 ,  G06V10/82 ,  G06V20/70 ,  G06V40/171 ,  G06T2207/30201 ,  G06V2201/07

    Abstract: Landmark detection refers to the detection of landmarks within an image or a video, and is used in many computer vision tasks such emotion recognition, face identity verification, hand tracking, gesture recognition, and eye gaze tracking. Current landmark detection methods rely on a cascaded computation through cascaded networks or an ensemble of multiple models, which starts with an initial guess of the landmarks and iteratively produces corrected landmarks which match the input more finely. However, the iterations required by current methods typically increase the training memory cost linearly, and do not have an obvious stopping criteria. Moreover, these methods tend to exhibit jitter in landmark detection results for video. The present disclosure improves current landmark detection methods by providing landmark detection using an iterative neural network. Furthermore, when detecting landmarks in video, the present disclosure provides for a reduction in jitter due to reuse of previous hidden states from previous frames.

    Method for Uncertainty Estimation in Object Detection Models
    360.
    发明公开
    Method for Uncertainty Estimation in Object Detection Models 审中-公开

    公开(公告)号：US20240095945A1

    公开(公告)日：2024-03-21

    申请号：US18464245

    申请日：2023-09-10

    Applicant: Aptiv Technologies (2) S.à r.l.

    Inventor： Weimeng Zhu

    IPC: G06T7/70 ,  G06T7/20 ,  G06T7/62 ,  G06V10/25 ,  G06V10/44

    CPC classification number: G06T7/70 ,  G06T7/20 ,  G06T7/62 ,  G06V10/25 ,  G06V10/44 ,  G06T2207/20081 ,  G06V2201/07

    Abstract: A computer-implemented method for evaluating a prediction quality of a model usable for detecting objects is disclosed. The method includes inputting, into the model, a set of data samples. Each data sample includes a scene representation including an object. The method includes outputting, by the model, a set of predictions. The set of predictions include, for each data sample of the set of data samples, a predicted feature of the object in the scene representation and a predicted uncertainty associated with the predicted feature. The method includes estimating an uncertainty estimation quality of the model based on the set of predictions. The method includes determining, based on the uncertainty estimation quality, whether a further training of the model to improve the uncertainty estimation quality is required.

Patent Agency Ranking