中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

22 records (took 0.02 seconds)

    Training a neural network to detect objects in images
    11.
    发明授权
    Training a neural network to detect objects in images 有权

    公开(公告)号：US09514389B1

    公开(公告)日：2016-12-06

    申请号：US15185613

    申请日：2016-06-17

    Applicant: Google Inc.

    Inventor： Dumitru Erhan ,  Christian Szegedy ,  Dragomir Anguelov

    IPC: G06K9/62 ,  G06K9/66

    CPC classification number: G06K9/6256 ,  G06K9/3241 ,  G06K9/4628 ,  G06K9/6202 ,  G06K9/66 ,  G06N3/0454 ,  G06N3/084

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network to detect object in images. One of the methods includes receiving a training image and object location data for the training image; providing the training image to a neural network and obtaining bounding box data for the training image from the neural network, wherein the bounding box data comprises data defining a plurality of candidate bounding boxes in the training image and a respective confidence score for each candidate bounding box in the training image; determining an optimal set of assignments using the object location data for the training image and the bounding box data for the training image, wherein the optimal set of assignments assigns a respective candidate bounding box to each of the object locations; and training the neural network on the training image using the optimal set of assignments.

    Estimation of panoramic camera orientation relative to a vehicle coordinate frame
    13.
    发明授权
    Estimation of panoramic camera orientation relative to a vehicle coordinate frame 有权
    估计相对于车辆坐标系的全景照相机方向

    公开(公告)号：US09270891B2

    公开(公告)日：2016-02-23

    申请号：US14206577

    申请日：2014-03-12

    Applicant: Google Inc.

    Inventor： Dragomir Anguelov ,  Daniel Joseph Filip

    IPC: H04N5/232 ,  H04N5/262 ,  G06T7/00

    CPC classification number: H04N5/23296 ,  G06T7/70 ,  G06T2207/30252 ,  H04N5/23238 ,  H04N5/2628

    Abstract: A system and method are presented for estimating the orientation of a panoramic camera mounted on a vehicle relative to the vehicle coordinate frame. An initial pose estimate of the vehicle is determined based on global positioning system data, inertial measurement unit data, and wheel odometry data of the vehicle. Image data from images captured by the camera is processed to obtain one or more tracks, each track including a sequence of matched feature points stemming from a same three-dimensional location. A correction parameter determined from the initial pose estimate and tracks can then be used to correct the orientations of the images captured by the camera. The correction parameter can be optimized by deriving a correction parameter for each of a multitude of distinct subsequences of one or more runs. Statistical analysis can be performed on the determined correction parameters to produce robust estimates.

    Abstract translation: 提出了一种用于估计安装在车辆上的全景相机相对于车辆坐标系的方位的系统和方法。 基于车辆的全球定位系统数据，惯性测量单元数据和车轮测距数据来确定车辆的初始姿态估计。 处理由相机拍摄的图像的图像数据以获得一个或多个轨道，每个轨道包括源自相同三维位置的匹配特征点的序列。 然后可以使用从初始姿态估计和轨迹确定的校正参数来校正由照相机拍摄的图像的取向。 可以通过为一个或多个运行的多个不同子序列中的每一个导出校正参数来优化校正参数。 可以对所确定的校正参数进行统计分析以产生鲁棒的估计。

    OBJECT DETECTION IN IMAGES BASED ON AFFINITY DETERMINATIONS
    14.
    发明申请
    OBJECT DETECTION IN IMAGES BASED ON AFFINITY DETERMINATIONS 有权
    基于AFFINITY决定的图像中的对象检测

    公开(公告)号：US20150169998A1

    公开(公告)日：2015-06-18

    申请号：US14204407

    申请日：2014-03-11

    Applicant: Google Inc.

    Inventor： Andrew Rabinovich ,  Dragomir Anguelov

    IPC: G06K9/62 ,  G06K9/66

    CPC classification number: G06K9/6218 ,  G06K9/00677 ,  G06K9/6219 ,  G06K9/6267 ,  G06K9/66

    Abstract: A hierarchy of clusters is determined, where each leave of the hierarchy corresponds to one of the images in a group, and each cluster in the hierarchy identifies images in the group that are deemed similar to one another. The hierarchy identifies a similarity between each of the plurality of clusters.

    Abstract translation: 确定集群的层次结构，其中层次结构的每个离开对应于组中的一个图像，并且层次中的每个集合标识组中认为彼此相似的图像。 该层次结构标识多个聚类中的每一个之间的相似性。

    Automatic video and dense image-based geographic information matching and browsing
    15.
    发明授权
    Automatic video and dense image-based geographic information matching and browsing 有权
    自动视频和密集的基于图像的地理信息匹配和浏览

    公开(公告)号：US08847951B1

    公开(公告)日：2014-09-30

    申请号：US14055242

    申请日：2013-10-16

    Applicant: Google Inc.

    Inventor： Dragomir Anguelov ,  Abhijit S. Ogale ,  Ehud Rivlin ,  Jay Yagnik

    IPC: G06T15/00 ,  G09G5/377

    CPC classification number: G09G5/377 ,  G06F3/1462 ,  G09G2340/12 ,  G09G2340/14 ,  G09G2354/00 ,  G09G2370/022 ,  G09G2380/00

    Abstract: Methods and systems permit automatic matching of videos with images from dense image-based geographic information systems. In some embodiments, video data including image frames is accessed. The video data may be segmented to determine a first image frame of a segment of the video data. Data representing information from the first image frame may be automatically compared with data representing information from a plurality of image frames of an image-based geographic information data system. Such a comparison may, for example, involve a search for a best match between geometric features, histograms, color data, texture data, etc. of the compared images. Based on the automatic comparing, an association between the video and one or more images of the image-based geographic information data system may be generated. The association may represent a geographic correlation between selected images of the system and the video data.

    Abstract translation: 方法和系统允许视频与来自基于图像的地理信息系统的图像自动匹配。 在一些实施例中，访问包括图像帧的视频数据。 视频数据可以被分割以确定视频数据的片段的第一图像帧。 表示来自第一图像帧的信息的数据可以与表示来自基于图像的地理信息数据系统的多个图像帧的信息的数据自动进行比较。 这样的比较可以例如涉及搜索比较图像的几何特征，直方图，颜色数据，纹理数据等之间的最佳匹配。 基于自动比较，可以生成视频与基于图像的地理信息数据系统的一个或多个图像之间的关联。 该关联可以表示系统的所选图像与视频数据之间的地理相关性。

    Automatic correction of trajectory data
    16.
    发明授权
    Automatic correction of trajectory data 有权
    自动校正轨迹数据

    公开(公告)号：US08831877B2

    公开(公告)日：2014-09-09

    申请号：US13965150

    申请日：2013-08-12

    Applicant: Google Inc.

    Inventor： Dragomir Anguelov ,  Scott Lininger ,  Ehud Rivlin

    IPC: G05D1/00 ,  G01C21/32

    CPC classification number: G01C21/32

    Abstract: Initial trajectory data that provides an initial description of an approximate trajectory of a device during a time period, and correction data that indicates a location of the device outside the approximate trajectory of the device within the time period, are received. A modified trajectory data that provides a modified description of a corrected trajectory of the device is generated. In particular, terms to express (i) location constraints that limit deformation of the approximate trajectory of the device and (ii) a modification constraint that limits departure of the corrected trajectory of the device from the location indicated by the correction data are generated, and the initial description of the approximate trajectory of the device is modified using the generated terms.

    Abstract translation: 初始轨迹数据，其提供在一段时间段期间设备的近似轨迹的初始描述，以及在该时间段内指示设备的大致轨迹外的设备的位置的校正数据。 生成修改后的轨迹数据，该轨迹数据提供对该装置的校正轨迹的修改的描述。 特别地，表达（i）限制装置的近似轨迹的变形的位置限制的术语和（ii）限制装置的校正轨迹从由校正数据指示的位置的偏离的修改约束，并且 使用生成的术语修改设备的近似轨迹的初始描述。

    Training a neural network to detect objects in images
    17.
    发明授权
    Training a neural network to detect objects in images 有权
    训练一个神经网络来检测图像中的物体

    公开(公告)号：US09373057B1

    公开(公告)日：2016-06-21

    申请号：US14528815

    申请日：2014-10-30

    Applicant: Google Inc.

    Inventor： Dumitru Erhan ,  Christian Szegedy ,  Dragomir Anguelov

    IPC: G06K9/00 ,  G06K9/62 ,  G06K9/66

    CPC classification number: G06K9/6256 ,  G06K9/3241 ,  G06K9/4628 ,  G06K9/6202 ,  G06K9/66 ,  G06N3/0454 ,  G06N3/084

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network to detect object in images. One of the methods includes receiving a training image and object location data for the training image; providing the training image to a neural network and obtaining bounding box data for the training image from the neural network, wherein the bounding box data comprises data defining a plurality of candidate bounding boxes in the training image and a respective confidence score for each candidate bounding box in the training image; determining an optimal set of assignments using the object location data for the training image and the bounding box data for the training image, wherein the optimal set of assignments assigns a respective candidate bounding box to each of the object locations; and training the neural network on the training image using the optimal set of assignments.

    Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于训练神经网络以检测图像中的对象。 其中一种方法包括接收训练图像和训练图像的对象位置数据; 将训练图像提供给神经网络并从神经网络获得用于训练图像的边界框数据，其中边界框数据包括定义训练图像中的多个候选边界框的数据和每个候选边界框的相应置信度分数 在训练形象中; 使用所述训练图像的所述对象位置数据和所述训练图像的所述边界框数据来确定最佳的分配集合，其中所述最佳分配集合将相应的候选边界框分配给每个所述对象位置; 并使用最佳赋值集在训练图像上训练神经网络。

    Object detection in images based on affinity determinations
    18.
    发明授权
    Object detection in images based on affinity determinations 有权
    基于亲和力测定的图像中的物体检测

    公开(公告)号：US09177226B2

    公开(公告)日：2015-11-03

    申请号：US14204407

    申请日：2014-03-11

    Applicant: Google Inc.

    Inventor： Andrew Rabinovich ,  Dragomir Anguelov

    IPC: G06K9/62 ,  G06K9/66

    CPC classification number: G06K9/6218 ,  G06K9/00677 ,  G06K9/6219 ,  G06K9/6267 ,  G06K9/66

    Abstract: A hierarchy of clusters is determined, where each leave of the hierarchy corresponds to one of the images in a group, and each cluster in the hierarchy identifies images in the group that are deemed similar to one another. The hierarchy identifies a similarity between each of the plurality of clusters.

    Abstract translation: 确定集群的层次结构，其中层次结构的每个离开对应于组中的一个图像，并且层次中的每个集合标识组中认为彼此相似的图像。 该层次结构标识多个聚类中的每一个之间的相似性。

Patent Agency Ranking