利索-全球专利检索

  1. Login
  2. Sign Up

Search Results

213 records (took 0.019 seconds)

    GAZE-TRACKING DEVICE, PROGRAM, AND METHOD
    1.
    发明公开
    GAZE-TRACKING DEVICE, PROGRAM, AND METHOD 审中-公开

    公开(公告)号：EP3413234A1

    公开(公告)日：2018-12-12

    申请号：EP18176601.5

    申请日：2018-06-07

    Applicant: AISIN SEIKI KABUSHIKI KAISHA

    Inventor： KOJIMA, Shinichi ,  OSUGA, Shin

    IPC: G06K9/00

    CPC classification number: G06T7/75 ,  G06K9/00268 ,  G06K9/0061 ,  G06T3/0068 ,  G06T2207/30201

    Abstract: A corneal-reflection-based gaze detection section (28) calculates a time series of a three-dimensional gaze vector in a camera coordinate system from a time series of facial images. A face position-and-orientation estimation section (24) estimates a time series of a three-dimensional position and orientation of a face. An eyeball-center-coordinates transformation section (32) calculates a time series of a three-dimensional position of the eyeball center in a coordinate system of a three-dimensional facial model. A fixed parameter calculation section (33) calculates for use as a fixed parameter a three-dimensional position of the eyeball center in the three-dimensional facial-model coordinate system. An eyeball-center-based gaze detection section (36) uses the three-dimensional position of the eyeball center calculated by the fixed parameter calculation section (33) to calculate a three-dimensional gaze vector from a three-dimensional position of the eyeball center to a three-dimensional position of a pupil center in the camera coordinate system. This enables accurate gaze tracking to be performed using a simple configuration and without performing calibration.

    METHOD AND DEVICE FOR MAGNETIC RESONANCE IMAGING WITH IMPROVED SENSITIVITY BY NOISE REDUCTION
    3.
    发明授权
    METHOD AND DEVICE FOR MAGNETIC RESONANCE IMAGING WITH IMPROVED SENSITIVITY BY NOISE REDUCTION 有权

    公开(公告)号：EP3295202B1

    公开(公告)日：2018-07-25

    申请号：EP15720915.6

    申请日：2015-05-08

    Applicant: Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.V.

    Inventor： MARSCHNER, Henrik ,  PAMPEL, André ,  MÖLLER, Harald

    IPC: G01R33/56 ,  G06T5/10 ,  G06T5/00

    CPC classification number: G06T11/005 ,  A61B5/055 ,  A61B5/7203 ,  G01R33/5608 ,  G01R33/56545 ,  G06T3/0068 ,  G06T5/002 ,  G06T5/10 ,  G06T2207/10088 ,  G06T2207/20064 ,  G06T2207/30016

    Abstract: A method of image processing of magnetic resonance (MR) images for creating de-noised MR images, comprises the steps of providing image data sets including multiple complex MR images (S7), subjecting the MR images to a wavelet decomposition (S12) for creating coefficient data sets of wavelet coefficients (Sn,m) representing the MR images in a wavelet frequency domain, calculating normalized coefficient data sets of wavelet coefficients Formula (I) (S17), wherein the coefficient data sets are normalized with a quantitative amount of variation, in particular standard deviation Formula (II), of noise contributions included in the coefficient data sets (Sn,m), averaging the wavelet coefficients of each coefficient data set (S18) for providing averaged wavelet coefficients Formula (III) of the coefficient data sets, calculating phase difference maps (Δφn,m) for all coefficient data sets (S20), wherein the phase difference maps provide phase differences between the phase of each wavelet coefficient and the phase of the averaged wavelet coefficients Formula (III), calculating scaled averaged coefficient data sets of wavelet coefficients by scaling the averaged wavelet coefficients Formula (III) with scaling factors (Cn,m), which are obtained by comparing parts of the normalized wavelet coefficients of the normalized coefficient data sets Formula (I) that are in phase with the averaged wavelet coefficients Formula (III) (S22), calculating rescaled coefficient data sets of wavelet coefficients Formula (IV) (S24) by applying a transfer function Formula (V) on the coefficient data sets (Sn,m) and on the scaled averaged coefficient data sets, wherein the transfer function includes combined amplitude and phase filters, each depending on the normalized coefficient data sets Formula (I) and me phase difference maps (Δφn,m), resp., and subjecting the rescaled coefficient data sets to a wavelet reconstruction Formula (IV) (S25) for providing the denoised MR images.

    METHOD FOR ALIGNMENT OF LOW-QUALITY NOISY DEPTH MAP TO THE HIGH-RESOLUTION COLOUR IMAGE
    6.
    发明公开
    METHOD FOR ALIGNMENT OF LOW-QUALITY NOISY DEPTH MAP TO THE HIGH-RESOLUTION COLOUR IMAGE 审中-公开
    将低质量噪声深度图与高分辨率彩色图像对齐的方法

    公开(公告)号：EP3213512A1

    公开(公告)日：2017-09-06

    申请号：EP15855684.5

    申请日：2015-10-23

    Applicant: Nokia Technologies Oy

    Inventor： SMIRNOV, Sergey ,  GEORGIEV, Mihail ,  GOTCHEV, Atanas

    IPC: H04N13/00 ,  G06T7/00 ,  G06T3/00

    CPC classification number: G06T15/205 ,  G06T3/0031 ,  G06T3/0068 ,  G06T3/4007 ,  G06T5/001 ,  G06T5/004 ,  G06T5/50 ,  G06T7/521 ,  G06T7/536 ,  G06T7/557 ,  G06T7/593 ,  G06T7/77 ,  G06T2207/10024 ,  G06T2207/10028 ,  G06T2207/10052 ,  G06T2207/20024 ,  G06T2207/20221 ,  G06T2207/30244 ,  G06T2215/16 ,  H04N13/111 ,  H04N13/128 ,  H04N13/189 ,  H04N13/271 ,  H04N13/282 ,  H04N2213/003

    Abstract: Various embodiments are provided which relate to the field of image signal processing, specifically relating to the generation of a depth-view image of a scene from a set of input images of a scene taken at different cameras of a multi-view imaging system. A method comprises obtaining a frame of an image of a scene and a frame of a depth map regarding the frame of the image. A minimum depth and a maximum depth of the scene and a number of depth layers for the depth map are determined. Pixels of the image are projected to the depth layers to obtain projected pixels on the depth layers; and cost values for the projected pixels are determined. The cost values are filtered and a filtered cost value is selected from a layer to obtain a depth value of a pixel of an estimated depth map.

    Abstract translation: 一种方法，装置或程序，用于获得场景的图像帧800和对应的深度图801，确定场景的最小和最大深度，确定深度图的深度层804的数量，投影图像的像素 确定投影像素的成本值828，对成本值824进行滤波，以及使用滤波后的成本值834获得像素的深度值。该实施例涉及生成高分辨率深度图像 使用高分辨率彩色摄像机拍摄的场景以及相关的低分辨率飞行时间深度传感器。 图像最好是彩色的。 成本值可以使用深度值置信度信息来计算。 其他独立权利要求不要求在计算层数之前确定最小或最大深度，而是根据估计的投影变换通过反投影来指定图像像素投影到深度层。 然而，更多独立的声明指出，获得场景的两幅图像以及关于两个不同观点的信息。 像素被投影并计算成本值并被过滤以从虚拟相机位置获得虚拟图像。

    REGISTERING FIRST IMAGE DATA OF A FIRST STREAM WITH SECOND IMAGE DATA OF A SECOND STREAM
    7.
    发明公开
    REGISTERING FIRST IMAGE DATA OF A FIRST STREAM WITH SECOND IMAGE DATA OF A SECOND STREAM 有权
    利用第二流的第二图像数据来注册第一流的第一图像数据

    公开(公告)号：EP3188120A1

    公开(公告)日：2017-07-05

    申请号：EP15202840.3

    申请日：2015-12-28

    Applicant: MedCom Gesellschaft für medizinische Bildverarbeitung mbH ,  Esaote S.p.A.

    Inventor： Sakas, Prof. Georgios ,  Kolev, Velizar ,  De Beni, Stefano ,  Forzoni, Leonardo

    IPC: G06T3/00 ,  G06T7/00 ,  A61B6/00

    CPC classification number: G06T7/33 ,  G06T3/0068 ,  G06T7/0012 ,  G06T7/38 ,  G06T2207/10016 ,  G06T2207/30004 ,  G06T2207/30048 ,  G06T2207/30061 ,  G06T2207/30196

    Abstract: The invention relates to a method of registering first image data of a first stream with second image data of a second stream, the first image data and the second image data comprising image information about a person or animal including image information about at least one organ (Or) of the person or animal, wherein a shape and/or position of the organ (Or) is changing thereby performing a cyclic change of the shape and/or position, and wherein:
    • each of the first stream and the second stream comprises a series of images (F) of the at least one organ (Or) at a plurality of different discrete stages of the cyclic change of the shape and/or position so that each of the images (F) corresponds to one of the discrete stages of the cyclic change of the shape and/or position,
    • a phase variable (p) is used to define the stage of the cyclic change of the shape and/or position and a phase value of the phase variable (p) is assigned to each of the images (F) within each of the first and second stream thereby defining unambiguously the discrete stage of the cyclic change of the shape and/or position represented by the image (F) instead of the time during a specific cycle when the image (F) was taken,
    • a registration, i.e. location-to-location associations of image values, is determined for a pair of images (F) consisting of a first image of the first stream and a second image of the second stream, wherein the first image and the second image of the pair are selected by choosing or interpolating an image from the first stream and by choosing or interpolating an image from the second stream to which the same phase value (pr) is assigned.

    Abstract translation: 本发明涉及一种将第一流的第一图像数据与第二流的第二图像数据进行配准的方法，第一图像数据和第二图像数据包括关于包括关于至少一个器官的图像信息的人或动物的图像信息 或者），其中所述器官（Or）的形状和/或位置正在改变，由此执行所述形状和/或位置的循环变化，并且其中：·所述第一流和所述第二流中的每一个包括 在所述形状和/或位置的循环变化的多个不同离散阶段处的所述至少一个器官（Or）的一系列图像（F），使得所述图像（F）中的每一个对应于所述离散阶段中的一个 •形状和/或位置的周期性变化•相位变量（p）用于定义形状和/或位置的循环变化阶段，并且相位变量（p）的相位值被分配给 第一个和第一个中的每一个内的图像（F）中的每一个 从而明确定义由图像（F）表示的形状和/或位置的周期性变化的离散阶段，而不是在拍摄图像（F）时的特定周期期间的时间，●配准，即位置 - 针对由第一流的第一图像和第二流的第二图像组成的一对图像（F）确定图像值的定位关联，其中该对的第一图像和第二图像由 选择或内插来自第一流的图像，以及通过选择或内插来自第二流的图像，其中分配了相同的相位值（pr）。

    INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM
    8.
    发明公开
    INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM 审中-公开
    信息管理部门，信息化计划

    公开(公告)号：EP3176755A1

    公开(公告)日：2017-06-07

    申请号：EP15826720.3

    申请日：2015-05-25

    Applicant: Sony Corporation

    Inventor： IGARASHI, Tatsuya ,  KIKKAWA, Norifumi

    IPC: G06T19/00 ,  G06F3/048

    CPC classification number: H04N5/44504 ,  G06F3/013 ,  G06F3/048 ,  G06T3/0068 ,  G06T19/00 ,  H04N5/23216 ,  H04N5/23293

    Abstract: The present disclosure is intended to display registration information on a real-world registration information setting point and update the display with detection of an event in accordance with a change in the context as a trigger. A control for displaying a virtual object including registration information on a real-world registration information setting point is performed. The data processing unit performs a process for displaying registration information described in control data and a process for updating the display in accordance with a change in predefined context. Further, the data processing unit displays the virtual object indicative of the registration information setting point in conjunction with the display of the registration information. The data processing unit calculates the positional relationship between an information processing apparatus and the registration information setting point on the basis of the position, direction, and the like of a reference image (anchor) detected from the shot image in a virtual three-dimensional space, and displays the virtual object indicative of the registration information setting point on the basis of the calculated positional relationship and a change in the context.

    Abstract translation: 本公开旨在在真实世界注册信息设置点上显示注册信息，并且根据上下文的改变作为触发来检测事件的更新显示。 执行用于在真实世界注册信息设置点上显示包括注册信息的虚拟对象的控件。 数据处理单元执行用于显示控制数据中描述的注册信息和根据预定上下文的变化更新显示的处理的处理。 此外，数据处理单元结合登记信息的显示显示指示登记信息设定点的虚拟对象。 数据处理单元基于在虚拟三维空间中从拍摄图像检测到的参考图像（锚）的位置，方向等来计算信息处理装置与登记信息设定点之间的位置关系 并且基于所计算的位置关系和上下文的改变来显示指示注册信息设置点的虚拟对象。

    METHOD AND APPARATUS FOR MULTIPLE IMAGE REGISTRATION IN THE GRADIENT DOMAIN
    10.
    发明公开
    METHOD AND APPARATUS FOR MULTIPLE IMAGE REGISTRATION IN THE GRADIENT DOMAIN 审中-公开
    VERFAHREN UND VORRICHTUNG ZUR REGISTRIERUNG MEHRERER BILDER IN DERGRADIENTENDOMÄNE

    公开(公告)号：EP3089109A1

    公开(公告)日：2016-11-02

    申请号：EP16158162.4

    申请日：2016-03-02

    Applicant: Nokia Technologies Oy

    Inventor： HUANG, Junzhou ,  CHEN, Xin

    IPC: G06T7/00 ,  G06T3/00

    CPC classification number: G06K9/6202 ,  G06T3/0068 ,  G06T7/30 ,  G06T7/32 ,  G06T7/33 ,  G06T11/60 ,  G06T2207/30184

    Abstract: A method, apparatus and computer program product are provided for image registration in the gradient domain. A method is provided including receiving three or more input images and registering, simultaneously, the three or more input images in the gradient domain based on applying an energy minimization function.

    Abstract translation: 提供了一种在梯度域中进行图像配准的方法，装置和计算机程序产品。 提供了一种方法，包括接收三个或更多个输入图像，并且基于应用能量最小化函数同时注册梯度域中的三个或更多个输入图像。

Patent Agency Ranking