中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

58 records (took 0.019 seconds)

    IMAGE-BASED POSE ESTIMATION AND ACTION DETECTION METHOD AND APPARATUS
    1.
    发明公开
    IMAGE-BASED POSE ESTIMATION AND ACTION DETECTION METHOD AND APPARATUS 审中-公开

    公开(公告)号：US20240046701A1

    公开(公告)日：2024-02-08

    申请号：US18088676

    申请日：2022-12-26

    Applicant: MARKANY INC.

    Inventor： Min Seong PARK ,  Chang Dae SON ,  Si Ye JANG ,  Ji In NAM

    IPC: G06V40/20 ,  G06V20/52 ,  H04N7/18 ,  G06V10/82

    CPC classification number: G06V40/20 ,  G06V20/52 ,  H04N7/183 ,  G06V10/82 ,  G08B13/19613

    Abstract: The present disclosure relates to a method of identifying a posture and detecting a specific behavior based on artificial intelligence. A method of detecting an abnormal behavior in a video based on a computational device according to an embodiment of the present disclosure may involve obtaining at least one video frame; obtaining at least one piece of human posture information from a first artificial intelligence based on the obtained video frame; obtaining information on whether an abnormal behavior has been detected and at least one piece of abnormal behavior information from a second artificial intelligence based on at least one piece of human posture information obtained in chronological order; and marking the at least one video frame based on the information on whether an abnormal behavior has been detected and the at least one piece of abnormal behavior information.

    METHOD AND DEVICE FOR AUGMENTING TRAINING DATA BY COMBINING OBJECT AND BACKGROUND
    2.
    发明申请
    METHOD AND DEVICE FOR AUGMENTING TRAINING DATA BY COMBINING OBJECT AND BACKGROUND 有权

    公开(公告)号：US20220207294A1

    公开(公告)日：2022-06-30

    申请号：US17137178

    申请日：2020-12-29

    Applicant: Markany Inc.

    Inventor： So Won KIM ,  Sae Yun JEON ,  Seung Yeob CHAE

    IPC: G06K9/62 ,  G06T11/60 ,  G06K9/00 ,  G06N3/08

    Abstract: Disclosed a method for augmenting training data by combining an object and a background with each other. The method includes extracting an object image, wherein the object image is a machine learning target; determining a type of the object image; receiving a background image, wherein the background image comprises a plurality of different background regions; identifying a first background region and a second background region among the plurality of different background regions; and combining the object image with the first background region and the second background region to augment training data, wherein combining the object image with the first background region and the second background region includes randomly positioning an image of a first type object corresponding to the first background region into the first background region, and randomly positioning an image of a second type object corresponding to the second background region into the second background region.

    Apparatus, system, and method of preventing forgery or falsification of electronic document based on content
    5.
    发明授权
    Apparatus, system, and method of preventing forgery or falsification of electronic document based on content 有权

    公开(公告)号：US10289812B2

    公开(公告)日：2019-05-14

    申请号：US15563537

    申请日：2017-09-07

    Applicant: MARKANY INC.

    Inventor： Dong-Hwa Kim ,  Yong-Chan Ahn ,  Ki-Soo Park

    IPC: G06F21/10 ,  H04L9/32 ,  G06F21/60 ,  G06K19/06 ,  G06F21/64 ,  G06F17/21 ,  G06Q50/10 ,  H04L12/22

    Abstract: The present invention includes: an electronic document receiving unit receiving an original electronic document; a text information extracting unit extracting text and text location information by analyzing content of the original electronic document; an image information extracting unit extracting an image and image location information by analyzing the content; a verification data generating unit generating original forgery falsification verification data by using at least one of the text and the text location information, the image and the image location information; and a secure electronic document generating unit generating a secure original electronic document after encrypting and inserting the original forgery falsification verification data in a preset position of the original electronic document. According to the present invention, a location of forged or falsified content can be indicated and identified with integrity verification of an electronic document file, whereby forgery or falsification can be effectively prevented.

    METHOD AND APPARATUS FOR EMBEDDING AND EXTRACTING TEXT WATERMARK
    6.
    发明申请
    METHOD AND APPARATUS FOR EMBEDDING AND EXTRACTING TEXT WATERMARK 审中-公开

    公开(公告)号：US20170329943A1

    公开(公告)日：2017-11-16

    申请号：US15593264

    申请日：2017-05-11

    Applicant: MARKANY INC.

    Inventor： Jonguk CHOI ,  Donghwa KIM ,  Jiseop MOON

    IPC: G06F21/16 ,  G06F21/60

    Abstract: One aspect of the present invention discloses a watermark insertion method. The method includes: segmenting target text into pieces of page content; obtaining a watermark variable comprising a line alternation value indicative of a watermark mode changed for each line of the segmented page content and a watermark mode setting value; and applying a flip-flop component insertion algorithm for inserting a watermark into each of the pieces of segmented page content based on the obtained watermark variable.

    Weight initialization method and apparatus for stable learning of deep learning model using activation function
    9.
    发明授权
    Weight initialization method and apparatus for stable learning of deep learning model using activation function 有权

    公开(公告)号：US11580406B2

    公开(公告)日：2023-02-14

    申请号：US16729506

    申请日：2019-12-30

    Applicant: Markany Inc.

    Inventor： Seung Yeob Chae ,  So Won Kim ,  Min Soo Park

    IPC: G06N3/08 ,  G06N3/084 ,  G06N3/04

    Abstract: Provided is an artificial neural network learning apparatus for deep learning. The apparatus includes an input unit configured to acquire an input data or a training data, a memory configured to store the input data, the training data, and a deep learning artificial neural network model, and a processor configured to perform computation based on the artificial neural network model, in which the processor sets the initial weight depending on the number of nodes belonging to a first layer and the number of nodes belonging to a second layer of the artificial neural network model, and determines the initial weight by compensation by multiplying a standard deviation (σ) by a square root of a reciprocal of a probability of a normal probability distribution for a remaining section except for a section in which an output value of the activation function converges to a specific value.

Patent Agency Ranking