公开(公告)号：WO2015057098A1

公开(公告)日：2015-04-23

申请号：PCT/RU2013/000921

申请日：2013-10-18

Applicant: LSI CORPORATION ,  KHOLODENKO, Alexander Borisovich ,  PARKHOMENKO, Denis Vladimirovich ,  PETYUSHKO, Alexander Alexandrovich ,  PARFENOV, Denis Vasilievich ,  ZAICEVS, Deniss

Inventor： KHOLODENKO, Alexander Borisovich ,  PARKHOMENKO, Denis Vladimirovich ,  PETYUSHKO, Alexander Alexandrovich ,  PARFENOV, Denis Vasilievich ,  ZAICEVS, Deniss

IPC: G06T5/00 ,  G06T7/20 ,  G01S17/89

CPC classification number: G06T7/2086 ,  G01S7/497 ,  G01S17/36 ,  G01S17/89 ,  G06T7/269 ,  G06T7/285 ,  G06T2207/10028 ,  H04N13/204

Abstract: In one embodiment, an image processor is configured to obtain a plurality of phase images for each of first and second depth frames. For each of a plurality of pixels of a given one of the phase images of the first depth frame, the image processor determines an amount of movement of a point of an imaged scene between the pixel of the given phase image and a pixel of a corresponding phase image of the second depth frame, and adjusts pixel values of respective other phase images of the first depth frame based on the determined amount of movement. A motion compensated first depth image is generated utilizing the given phase image and the adjusted other phase images of the first depth frame. Movement of a point of the imaged scene is determined, for example, between pixels of respective n-th phase images of the first and second depth frames.

Abstract translation: 在一个实施例中，图像处理器被配置为获得用于第一和第二深度帧中的每一个的多个相位图像。 对于第一深度帧的给定一个相位图像的多个像素中的每个像素，图像处理器确定给定相位图像的像素与相应图像的像素之间的成像场景的点的移动量 并且基于所确定的移动量来调整第一深度帧的各个其他相位图像的像素值。 使用给定的相位图像和第一深度帧的经调整的其他相位图像来生成运动补偿的第一深度图像。 例如，在第一和第二深度帧的各个第n相图像的像素之间确定成像场景的点的移动。

公开(公告)号：WO2014154533A1

公开(公告)日：2014-10-02

申请号：PCT/EP2014/055415

申请日：2014-03-18

Applicant: THOMSON LICENSING

Inventor： SORGI, Lorenzo ,  JACHALSKY, Joern

IPC: G06T7/00

CPC classification number: G06T7/2086 ,  G06K9/4604 ,  G06T7/285 ,  G06T7/579 ,  G06T7/74 ,  G06T2207/10016

Abstract: A method for extracting keyframes from a sequence of frames for a computer vision application using structure from motion, the keyframes being a subset of representative frames from the complete sequence of frames, and an apparatus configured to perform the method are described. A subset selector (22) selects (10) a subset of keyframes that closely match a current camera position from already available keyframes. A determination unit (23) then determines (11) whether a current frame should be included a bundle adjustment keyframe set and/or a triangulation keyframe set.

Abstract translation: 描述了使用来自运动的结构的计算机视觉应用的帧序列中提取关键帧的方法，关键帧是来自完整帧序列的代表帧的子集，以及被配置为执行该方法的装置。 子集选择器（22）从已经可用的关键帧中选择（10）与当前摄像机位置紧密匹配的关键帧的子集。 确定单元（23）然后确定（11）是否应该包括当前帧，包束调整关键帧集合和/或三角关键帧集合。

公开(公告)号：WO2015188038A1

公开(公告)日：2015-12-10

申请号：PCT/US2015/034354

申请日：2015-06-05

Applicant: PROMETHEAN LIMITED ,  VELOZO, Steven C.

Inventor： VELOZO, Steven C. ,  DECKER, Alex, H. ,  HILLIER, Jason, C.

IPC: G06K9/00

CPC classification number: G06T7/2086 ,  G06K9/00771 ,  G06T7/285 ,  G06T7/292 ,  G06T7/66 ,  G06T2207/10016 ,  G06T2207/30232 ,  H04N7/181

Abstract: A system and method for tracking association of two or more objects over time, according to various embodiments, is configured to determine the association based at least in part on an image. The system may be configured to capture the image, identify two or more objects of interest within the image, determine whether the two or more objects are associated in the image, and store image association data for the two or more objects. In various embodiments the system is configured to create a timeline of object association over time for display to a user.

Abstract translation: 根据各种实施例，用于跟踪两个或多个对象随时间的关联的系统和方法被配置为至少部分地基于图像来确定关联。 该系统可以被配置为捕获图像，识别图像内的两个或更多个感兴趣对象，确定两个或更多个对象是否与图像相关联，并且存储用于两个或更多个对象的图像关联数据。 在各种实施例中，系统被配置为随时间创建对象关联的时间线以向用户显示。

公开(公告)号：WO2010037110A1

公开(公告)日：2010-04-01

申请号：PCT/US2009/058824

申请日：2009-09-29

Applicant: IMAGEMOVERS DIGITAL LLC ,  GRIFFIN, Doug

Inventor： GRIFFIN, Doug

IPC: G06T15/00

CPC classification number: G06T7/2086 ,  G06F3/011 ,  G06K9/00335 ,  G06T7/285 ,  G06T2200/08 ,  G06T2207/10021 ,  G06T2207/30201 ,  G06T2207/30204

Abstract: The present invention relates to computer capture of object motion. More specifically, embodiments of the present invention relate to capturing of facial movement or performance of an actor. Embodiments of the present invention provide a head-mounted camera system that allows the movements of an actor's face to be captured separately from, but simultaneously with, the movements of the actor's body. In some embodiments of the present invention, a method of motion capture of an actor's performance is provided. A self- contained system is provided for recording the data, which is free of tethers or other hard¬ wiring, is remotely operated by a motion-capture team, without any intervention by the actor wearing the device. Embodiments of the present invention also provide a method of validating that usable data is being acquired and recorded by the remote system.

Abstract translation: 本发明涉及物体运动的计算机捕获。 更具体地，本发明的实施例涉及捕获表情运动或表演的演员。 本发明的实施例提供一种头戴式摄像机系统，其允许演员的脸部的运动与演员的身体的运动分开地但同时被捕获。 在本发明的一些实施例中，提供了一种运动捕捉演员演奏的方法。 提供了一种独立的系统，用于记录没有系绳或其他硬接线的数据，由运动捕获团队远程操作，而不需要穿着设备的演员的干预。 本发明的实施例还提供一种验证由远程系统获取和记录可用数据的方法。

公开(公告)号：WO2014160267A2

公开(公告)日：2014-10-02

申请号：PCT/US2014026196

申请日：2014-03-13

Applicant: MICROSOFT CORP

Inventor： HOOF JONATHAN R ,  KENNETT DANIEL G

IPC: A63F13/213

CPC classification number: G06T7/2086 ,  A63F13/213 ,  A63F13/428 ,  G06F3/011 ,  G06F3/017 ,  G06T7/251 ,  G06T7/66 ,  G06T2200/04 ,  G06T2207/10016 ,  G06T2207/10028 ,  G06T2207/30196

Abstract: Techniques described herein determine a center of mass state vector based on a body model. The body model may be formed by analyzing a depth image of a user who is performing some motion. The center of mass state vector may include, for example, center-of-mass position, center-of-mass velocity, center-of-mass acceleration, orientation, angular velocity, angular acceleration, inertia tensor, and angular momentum. A center of mass state vector may be determined for an individual body part or for the body as a whole. The center of mass state vector(s) may be used to analyze the user's motion.

Abstract translation: 本文描述的技术基于身体模型确定质心状态向量的中心。 身体模型可以通过分析正在执行某些运动的用户的深度图像来形成。 质心状态向量的中心可以包括例如质心位置，质心速度，质心加速度，取向，角速度，角加速度，惯性张量和角动量。 可以针对个体身体部位或身体整体确定质量状态向量的中心。 质量状态矢量的中心可以用于分析用户的运动。

公开(公告)号：WO2014151700A1

公开(公告)日：2014-09-25

申请号：PCT/US2014/026270

申请日：2014-03-13

Applicant: MICROSOFT CORPORATION

Inventor： HOOF, Jonathan R. ,  KENNETT, Daniel G. ,  AHMAD, Anis

IPC: A63F13/20 ,  A63F13/42 ,  A63F13/213

CPC classification number: G06T7/2086 ,  A63F13/06 ,  A63F2300/1087 ,  A63F2300/6045 ,  A63F2300/69 ,  G06F3/017 ,  G06T7/285

Abstract: Techniques described herein use signal analysis to detect and analyze repetitive user motion that is captured in a 3D image. The repetitive motion could be the user exercising. One embodiment includes analyzing image data that tracks a user performing a repetitive motion to determine data points for a parameter that is associated with the repetitive motion. The different data points are for different points in time. A parameter signal of the parameter versus time that tracks the repetitive motion is formed. The parameter signal is divided into brackets that delineate one repetition of the repetitive motion from other repetitions of the repetitive motion. A repetition in the parameter signal is analyzed using a signal processing technique. Curve fitting and/or auto-correlation may be used to analyze the repetition.

Abstract translation: 本文描述的技术使用信号分析来检测和分析在3D图像中捕获的重复用户运动。 重复运动可能是用户行使。 一个实施例包括分析跟踪执行重复运动的用户的图像数据，以确定与重复运动相关联的参数的数据点。 不同的数据点是不同的时间点。 形成跟踪重复运动的参数对时间的参数信号。 参数信号被分成括号，其中描述了重复运动的一个重复与其他重复运动的重复。 使用信号处理技术分析参数信号中的重复。 可以使用曲线拟合和/或自相关来分析重复。

公开(公告)号：WO2011102871A1

公开(公告)日：2011-08-25

申请号：PCT/US2010/060745

申请日：2010-12-16

Applicant: PANASONIC CORPORATION ,  LEE, Kuo Chu ,  OZDEMIR, Hasan Timucin ,  SHI, Xiangiun ,  LIU, Lipin ,  JOO, Namsoo ,  YU, Juan ,  LIU, Jun

Inventor： LEE, Kuo Chu ,  OZDEMIR, Hasan Timucin ,  SHI, Xiangiun ,  LIU, Lipin ,  JOO, Namsoo ,  YU, Juan ,  LIU, Jun

IPC: G08B13/196 ,  H04N7/18

CPC classification number: G06T7/2086 ,  G06K9/00771 ,  G06T7/285 ,  G08B13/19608 ,  G08B13/19676

Abstract: A video surveillance system is disclosed. The system includes a model database storing a plurality of models and a vector database storing a plurality of vectors of recently observed trajectories. The system includes a model building module that builds a new motion model corresponding to the motion data of the current trajectory data structure. The system generates a current trajectory data structure having motion data and abnormality scores. The system also includes a database purging module configured to determine a subset of vectors that is most similar to the current trajectory data structure based on a measure of similarity between the subset of vectors and the current trajectory data structure. The database purging module is further configured to replace one of the motion models in the model database with the new motion model based on an amount of vectors in the subset vectors the recentness of the subset of vectors.

Abstract translation: 公开了一种视频监控系统。 该系统包括存储多个模型的模型数据库和存储最近观察到的轨迹的多个矢量的矢量数据库。 该系统包括建立与当前轨迹数据结构的运动数据相对应的新运动模型的模型构建模块。 该系统产生具有运动数据和异常分数的当前轨迹数据结构。 该系统还包括数据库清除模块，其被配置为基于矢量子集与当前轨迹数据结构之间的相似度测量来确定与当前轨迹数据结构最相似的向量子集。 数据库清除模块还被配置为基于子集向量中矢量子集的近似度，用新的运动模型替换模型数据库中的运动模型中的一个。

公开(公告)号：WO2017087398A1

公开(公告)日：2017-05-26

申请号：PCT/US2016/062059

申请日：2016-11-15

Applicant: ABB SCHWEIZ AG ,  WANG, Jianjun ,  ZHANG, Biao ,  MARTINEZ, Carlos ,  MORATO, Carlos, W. ,  BOCA, Remus

Inventor： WANG, Jianjun ,  ZHANG, Biao ,  MARTINEZ, Carlos ,  MORATO, Carlos, W. ,  BOCA, Remus

IPC: G05B15/00

CPC classification number: B25J9/1697 ,  B25J9/1692 ,  G06T7/002 ,  G06T7/004 ,  G06T7/2086 ,  G06T7/74 ,  G06T2200/04 ,  G06T2207/10028 ,  G06T2207/30244 ,  Y10S901/09

Abstract: Robot positioning is facilitated by obtaining, for each time of a first sampling schedule, a respective indication of a pose of a camera system of a robot relative to a reference coordinate frame, the respective indication of the pose of the camera system being based on a comparison of multiple three-dimensional images of a scene of an environment, the obtaining providing a plurality of indications of poses of the camera system; obtaining, for each time of a second sampling schedule, a respective indication of a pose of the robot, the obtaining providing a plurality of indications of poses of the robot; and determining, using the plurality of indications of poses of the camera system and the plurality of indications of poses of the robot, an indication of the reference coordinate frame and an indication of a reference point of the camera system relative to pose of the robot.

Abstract translation: 通过针对第一采样时间表的每一次获得机器人的相机系统相对于参考坐标系的姿势的相应指示来促进机器人定位，姿势的相应指示 所述相机系统的基于环境的场景的多个三维图像的比较的基础上，所述获得提供所述相机系统的姿势的多个指示; 针对第二采样时间表的每个时间，获得机器人的姿势的相应指示，所述获得提供机器人的姿势的多个指示; 以及使用所述相机系统的多个姿势指示和所述机器人的姿势的多个指示来确定所述参考坐标框的指示以及所述相机系统的参考点相对于所述机器人的姿势的指示。

公开(公告)号：WO2017087083A1

公开(公告)日：2017-05-26

申请号：PCT/US2016/055528

申请日：2016-10-05

Applicant: GOOGLE INC.

Inventor： DONNER, Craig ,  LALONDE, Paul Albert ,  PARKER, Evan Hardesty

IPC: G06F3/01 ,  G06F3/038 ,  H04N13/04 ,  G02B27/01 ,  G06F3/0346 ,  G02B27/00

CPC classification number: G06T15/04 ,  G02B27/0093 ,  G02B27/017 ,  G02B27/0172 ,  G02B27/0179 ,  G02B2027/0132 ,  G02B2027/014 ,  G02B2027/0187 ,  G06F3/011 ,  G06F3/012 ,  G06F3/0346 ,  G06F3/038 ,  G06T7/2086 ,  G06T7/285 ,  G06T7/408 ,  G06T7/90 ,  G06T19/006 ,  G06T2200/04 ,  H04N13/344 ,  H04N13/376

Abstract: A system (100) includes a head mounted display (HMD) device (102) comprising at least one display (112, 114) and at least one sensor (210) to provide pose information for the HMD device. The system further includes a sensor integrator module (304) coupled to the at least one sensor, the sensor integrator module to determine a motion vector for the HMD device based on the pose information, and an application processor (204) to render a first texture (122, 322) based on pose of the HMD device determined from the pose information. The system further includes a motion analysis module (302) to determine a first velocity field (128, 328) having a pixel velocity for at least a subset of pixels of the first texture, and a compositor (224) to render a second texture (140, 334) based on the first texture, the first velocity field and the motion vector for the HMD, and to provide the second texture to the display of the HMD device.

Abstract translation: 系统（100）包括包含至少一个显示器（112,114）和至少一个传感器（210）的头戴式显示器（HMD）设备（102），以提供HMD的姿态信息 设备。 该系统还包括耦合到至少一个传感器的传感器集成器模块（304），传感器集成器模块基于姿势信息确定HMD设备的运动矢量，以及应用处理器（204），用于渲染第一纹理 （122,322），其基于从姿势信息确定的HMD设备的姿态。 该系统还包括：运动分析模块（302），用于确定具有用于第一纹理的像素的至少一个子集的像素速度的第一速度场（128,328）;以及合成器（224），用于渲染第二纹理 基于第一纹理，第一速度场和用于HMD的运动矢量，并将第二纹理提供给HMD设备的显示器。

公开(公告)号：WO2011115635A1

公开(公告)日：2011-09-22

申请号：PCT/US2010/034706

申请日：2010-05-13

Applicant: UNIVERSITY OF CENTRAL FLORIDA RESEARCH FOUNDATION, INC. ,  GUPTA, Prince ,  DA VITORIA LOBO, Niels ,  LAVIOLA, Joseph, J., Jr.

Inventor： GUPTA, Prince ,  DA VITORIA LOBO, Niels ,  LAVIOLA, Joseph, J., Jr.

IPC: G06K9/00

CPC classification number: G06T7/2086 ,  G06T7/269 ,  G06T7/285

Abstract: Systems and method of compensating for tracking motion of an object are disclosed. One such method includes receiving a series of images captured by each of a plurality of image capture devices. The image capture devices are arranged in an orthogonal configuration of two opposing pairs. The method further includes computing a series of positions of the object and orientations of the object, by processing the images captured by each of the plurality of image capture devices.

Abstract translation: 公开了补偿对象的跟踪运动的系统和方法。 一种这样的方法包括接收由多个图像捕获设备中的每一个捕获的一系列图像。 图像捕获装置以两对相对的正交配置排列。 该方法还包括通过处理由多个图像捕获设备中的每一个捕获的图像来计算物体的一系列位置和对象的取向。