利索-全球专利检索

  1. Login
  2. Sign Up

Search Results

1254 records (took 0.026 seconds)

    PRIVACY-PRESERVING FACIAL SCREENING
    3.
    发明申请
    PRIVACY-PRESERVING FACIAL SCREENING 有权

    公开(公告)号：US20250047492A1

    公开(公告)日：2025-02-06

    申请号：US18791869

    申请日：2024-08-01

    Applicant: UT-Battelle, LLC

    Inventor： Suhas Sreehari ,  Hector Santos-Villalobos

    IPC: H04L9/32 ,  G06F9/451 ,  G06V10/774 ,  G06V10/82 ,  G06V10/94 ,  G06V40/16

    Abstract: Privacy-preserving security screening using facial images is disclosed. The screening may comprise encoding a facial image into a latent vector using a first pre-trained machine learning model, shuffling the latent vector using an invertible transformation in a latent space to generate a shuffled latent vector and converting the shuffled latent vector into a landscape image using a second pre-trained machine learning model. The landscape image may be used to security screening and compared with reference landscape images.

    INTELLIGENT DOORBELL FOR A SECURITY SYSTEM
    4.
    发明申请
    INTELLIGENT DOORBELL FOR A SECURITY SYSTEM 有权

    公开(公告)号：US20250046165A1

    公开(公告)日：2025-02-06

    申请号：US18340835

    申请日：2023-06-24

    Applicant: Khurram Hussain

    Inventor： Khurram Hussain

    IPC: G08B13/196 ,  G06V10/94 ,  G06V20/30 ,  G06V20/52 ,  G06V40/16 ,  G06V40/20 ,  H04L12/28

    Abstract: A smart doorbell, for a security system, which is capable of performing functions including, person detection, facial recognition, motion detection, detection of baby or pet running away, weapon detection, fire detection, detection of parcel left at porch, parcel theft, dog poop detection, vehicle detection, animal detection, eavesdropper detection, a person jumping from backyard wall, and intruder detection. The functions are detected through a camera of the doorbell and, subsequently, AI-based alerts are generated.

    Systems and methods for capturing data from a medical device
    5.
    发明授权
    Systems and methods for capturing data from a medical device 有权

    公开(公告)号：US12217850B2

    公开(公告)日：2025-02-04

    申请号：US17531322

    申请日：2021-11-19

    Applicant: ABIOMED, Inc.

    Inventor： Alessandro Simone Agnello

    IPC: G16H30/20 ,  G06V10/94 ,  G06V30/146 ,  G06V30/148 ,  G16H15/00 ,  G16H30/40 ,  G16H40/40 ,  A61M60/122 ,  G16H10/60

    Abstract: A method for transferring data from a medical device to a server comprises receiving a video stream from the medical device, capturing an image from the video stream, transmitting the image to the server via a data network, and extracting the data from the image. The image may illustrate and/or represent data over a period of time. The method may also comprise transmitting, from a data module receiving the video stream from the medical device, a signal to a router that indicates that the data module is connected to the network. The method may also comprise transmitting a command to the data module to start capturing the image, transferring the image to the router, broadcasting a signal indicating that the data module has captured the image, receiving the broadcasted signal at the server, and storing the image at the server.

    Method to facilitate mass conversion of 2D drawings to 3D models
    6.
    发明授权
    Method to facilitate mass conversion of 2D drawings to 3D models 有权

    公开(公告)号：US12217376B2

    公开(公告)日：2025-02-04

    申请号：US17871758

    申请日：2022-07-22

    Applicant: Draawn, LLC

    Inventor： James Cotteleer ,  Mark Cotteleer

    IPC: G06T19/20 ,  G06T7/00 ,  G06V10/94 ,  G06V20/30 ,  G06V20/64

    Abstract: An internet or cloud-based system, method, or platform (“platform”) used to facilitate the conversion of electronic two-dimensional drawings to three-dimensional models. A group of people (“crowd”) that has been found qualified to make such conversions, are selected for the conversion. The two-dimensional drawings are transmitted to the crowd for conversion to three-dimensional models. In some embodiments, multiple instances of the same two-dimensional drawings (or image data) is sent to multiple, independent crowd members in order that multiple versions of the same three-dimensional model can be created. Once the models are complete and returned, they are compared to each other on multiple features or characteristics. If two or more three-dimensional models are found to match within the prescribed tolerances, they are determined to be an accurate representation of the product or device shown in the two-dimensional drawings. In some embodiments, the two-dimensional drawings can be divided into subparts and submitted to different crowd members for conversion.

    SURVEILLANCE CONTAINER FOR VEHICLE ROOF
    7.
    发明申请
    SURVEILLANCE CONTAINER FOR VEHICLE ROOF 有权

    公开(公告)号：US20250029358A1

    公开(公告)日：2025-01-23

    申请号：US18355121

    申请日：2023-07-19

    Applicant: ASSURE MSP, Inc.

    Inventor： George FIDENTIZIS ,  James GRAF ,  Scott SNYDER

    IPC: G06V10/70 ,  G06V10/94 ,  G06V20/56 ,  H04N7/18

    Abstract: In some examples, a vehicle rooftop surveillance system may include a container. In such examples, the container may include a base comprising a top portion and a bottom portion, a base opening at the bottom portion of the base and an interior surface extending from the base opening to the top portion of the base. Additionally, the container may include a first viewing window adapted so that a first sensor within the container can detect objects outside the container. Further, the container may include a fastener element configured to attach the base to a roof of a vehicle.

    Computer vision model drawing interface
    8.
    发明授权
    Computer vision model drawing interface 有权

    公开(公告)号：US12205363B2

    公开(公告)日：2025-01-21

    申请号：US17817611

    申请日：2022-08-04

    Applicant: Rockwell Automation Technologies, Inc.

    Inventor： Alexander Berman ,  Bijan SayyarRodsari ,  Wei Dai ,  Kadir Liano

    IPC: G06V10/00 ,  G06V10/22 ,  G06V10/75 ,  G06V10/774 ,  G06V10/776 ,  G06V10/94 ,  G06V10/82

    Abstract: For updating a computer vision model, a method converts a user input drawing including a user annotation of a first image in a drawing format to a training format image in a training format for a computer vision model. The method generates a training-representation drawing from the training format image. The training-representation drawing includes an image inference for the first image. The method receives user feedback for the training-representation drawing in the drawing format. The method updates the computer vision model based on the user feedback. The method generates an image inference for a second image based on the updated computer vision model and generates model-health metrics, agreement-metrics, and a sortable image index to explain image inferences with respect to guided user annotation of the second image. The method caches partial results from the image-inferences to afford quicker updating of computer vision models, affording more iterative model-development than ad-hoc model-evaluation.

Patent Agency Ranking