公开(公告)号：US09437004B2

公开(公告)日：2016-09-06

申请号：US14311989

申请日：2014-06-23

Applicant: Google Inc.

Inventor： Jonathan Siegel ,  Allen Hutchison ,  Wenfeng Li

IPC: G06T7/00

CPC classification number: G06T7/0044 ,  G06T7/254 ,  G06T2200/04 ,  G06T2207/10028 ,  G06T2207/10032 ,  G06T2207/20021 ,  G06T2207/30184

Abstract: An exemplary method for surfacing feature changes occurring over time at a location is provided. The method includes identifying sets of images captured at a location over a period of time. The sets of images depict objects at the location. Visual changes at the location over the period of time are determined for a given physical feature of the objects depicted in images from the sets of images based on at least a comparison of features in corresponding areas of a first and a second 3D geometry of the location associated with the images. The first and second 3D geometries depict physical features of objects at the location at different times. The sets of images may be ranked based on the determined visual changes for each set. A given set of images from the sets of images may be selected for display on a user interface based on the ranking.

Abstract translation: 提供了一种用于表面处理的示例性方法，其特征在于在某个位置随时间发生的变 该方法包括识别在一段时间内在某个位置拍摄的图像集合。 该组图像描绘了位置处的对象。 基于至少基于位置的第一和第二3D几何的相应区域中的特征的比较来确定在一段时间内的位置处的视觉变化，用于来自图像集合的图像中描绘的对象的给定物理特征 与图像相关联。 第一和第二3D几何图形描绘了不同时间位置处物体的物理特征。 可以基于每组的确定的视觉变化对图像集进行排名。 可以选择来自该组图像的一组给定的图像，以便基于排名在用户界面上显示。

公开(公告)号：US20140297575A1

公开(公告)日：2014-10-02

申请号：US13854314

申请日：2013-04-01

Applicant: GOOGLE INC.

Inventor： Evan Rapoport ,  Allen Hutchison ,  Kei Kawai ,  Luc Vincent

IPC: G06N5/04

CPC classification number: G06N5/04 ,  G06F16/29 ,  G06F16/58 ,  G06F16/9537

Abstract: Systems and methods of the present disclosure provide techniques for providing user-specified ways of navigating through real-world three-dimensional geographic imagery that spans space and time. An exemplary method includes identifying a plurality of images depicting a geographic location at street level. The images are captured at the geographic location over a span of time. Using a processor, image data is associated with the plurality of images. The image data includes information representing positional data and a time dimension related to the plurality of images. Using the processor, a user's navigational intent to move back and forward through the time dimension is predicted based on a navigational signal. The exemplary method further includes selecting a set of images from the plurality of images based on the image data and the predicted navigational intent. The set of images depict conditions at the geolocation for one or more time periods.

Abstract translation: 本公开的系统和方法提供了用于提供用户指定的通过跨越空间和时间的真实世界三维地理图像导航的方法的技术。 示例性方法包括识别描绘街道级别的地理位置的多个图像。 这些图像在一段时间内在地理位置被捕获。 使用处理器，图像数据与多个图像相关联。 图像数据包括表示与多个图像相关的位置数据和时间维度的信息。 使用处理器，基于导航信号预测用户在时间维度上前后移动的导航意图。 该示例性方法还包括基于图像数据和预测的导航意图从多个图像中选择一组图像。 这组图像描绘了一个或多个时间段的地理位置条件。

公开(公告)号：US09280820B2

公开(公告)日：2016-03-08

申请号：US14337680

申请日：2014-07-22

Applicant: Google Inc.

Inventor： Allen Hutchison

IPC: G06T7/00 ,  G06K9/62

CPC classification number: G06K9/6202 ,  G06K9/00664 ,  G06K9/6267 ,  H04N5/23229

Abstract: Systems and methods are provided for using imagery depicting a timekeeping device to determine a clock offset for a particular image capture device. The clock offset can be used to correct timestamps associated with one or more images captured by such image capture device. One example method includes analyzing imagery depicting at least in part a timekeeping device to determine a first time displayed by the timekeeping device in the imagery. The method includes determining whether the first time comprises a 12-hour value or a 24-hour value. The method includes, when it is determined that the first time comprises a 12-hour value, determining a corresponding 24-hour value for the 12-hour value based at least in part on information contained within a plurality of images. The method includes determining a clock offset between the 24-hour value and the first timestamp. One example system includes a timestamp correction engine for correcting timestamps.

Abstract translation: 提供的系统和方法用于使用描绘计时装置的图像来确定特定图像捕获装置的时钟偏移。 时钟偏移可用于校正与由这种图像捕获设备捕获的一个或多个图像相关联的时间戳。 一个示例性方法包括分析至少部分地描绘计时装置以确定由图像中的计时装置显示的第一次的图像。 该方法包括确定第一次是否包括12小时值或24小时值。 该方法包括：当确定第一次包括12小时值时，至少部分地基于包含在多个图像中的信息来确定对于12小时值的相应24小时值。 该方法包括确定24小时值和第一时间戳之间的时钟偏移。 一个示例系统包括用于校正时间戳的时间戳校正引擎。

公开(公告)号：US20150302633A1

公开(公告)日：2015-10-22

申请号：US14258594

申请日：2014-04-22

Applicant: Google Inc.

Inventor： Wenfeng Li ,  Allen Hutchison

IPC: G06T15/10 ,  G06T3/00

CPC classification number: G06T15/10 ,  G06T3/0012 ,  G09G5/14 ,  G09G2340/045 ,  G09G2360/122 ,  G09G2380/10

Abstract: Images may be automatically annotated with place identifiers based on textual metadata associated with the images. The quality of these place identifiers may be scored based on a number of different factors. For example, these annotations may also include confidence and topicality values indicative of the relationships between the place identifier and the text as well as topics of the text of the textual metadata. Images that also have additional location information that includes GPS coordinates or a location specified by a user, may be used to evaluate the quality of a given place identifier, combination of confidence and topicality values, and the overall annotation system. These valuations may, in turn, be used to score the quality of such automatic annotations of other images that do not have such additional location information.

Abstract translation: 可以使用基于与图像相关联的文本元数据的地点标识符来自动注释图像。 这些地点标识符的质量可以基于许多不同的因素进行评分。 例如，这些注释还可以包括指示地点标识符和文本之间的关系以及文本元数据的文本的主题的置信度和话题值。 还可以使用还具有包括GPS坐标或用户指定的位置的附加位置信息的图像来评估给定地点标识符的质量，置信度和话题值的组合以及总体注释系统。 这些估值又可以用于评估其他图像的自动注释的质量，这些图像不具有这样的附加位置信息。

公开(公告)号：US09972121B2

公开(公告)日：2018-05-15

申请号：US14258594

申请日：2014-04-22

Applicant: Google Inc.

Inventor： Wenfeng Li ,  Allen Hutchison

IPC: G06T15/00 ,  G06T15/10 ,  G06T3/00 ,  G09G5/14

CPC classification number: G06T15/10 ,  G06T3/0012 ,  G09G5/14 ,  G09G2340/045 ,  G09G2360/122 ,  G09G2380/10

Abstract: Images may be automatically annotated with place identifiers based on textual metadata associated with the images. The quality of these place identifiers may be scored based on a number of different factors. For example, these annotations may also include confidence and topicality values indicative of the relationships between the place identifier and the text as well as topics of the text of the textual metadata. Images that also have additional location information that includes GPS coordinates or a location specified by a user, may be used to evaluate the quality of a given place identifier, combination of confidence and topicality values, and the overall annotation system. These valuations may, in turn, be used to score the quality of such automatic annotations of other images that do not have such additional location information.

公开(公告)号：US20150371389A1

公开(公告)日：2015-12-24

申请号：US14311989

申请日：2014-06-23

Applicant: GOOGLE INC.

Inventor： Jonathan Siegel ,  Allen Hutchison ,  Wenfeng Li

IPC: G06T7/00

CPC classification number: G06T7/0044 ,  G06T7/254 ,  G06T2200/04 ,  G06T2207/10028 ,  G06T2207/10032 ,  G06T2207/20021 ,  G06T2207/30184

Abstract: An exemplary method for surfacing feature changes occurring over time at a location is provided. The method includes identifying sets of images captured at a location over a period of time. The sets of images depict objects at the location. Visual changes at the location over the period of time are determined for a given physical feature of the objects depicted in images from the sets of images based on at least a comparison of features in corresponding areas of a first and a second 3D geometry of the location associated with the images. The first and second 3D geometries depict physical features of objects at the location at different times. The sets of images may be ranked based on the determined visual changes for each set. A given set of images from the sets of images may be selected for display on a user interface based on the ranking.

Abstract translation: 提供了一种用于表面处理的示例性方法，其特征在于在某个位置随时间发生的变 该方法包括识别在一段时间内在某个位置拍摄的图像集合。 该组图像描绘了位置处的对象。 基于至少基于位置的第一和第二3D几何的相应区域中的特征的比较来确定在一段时间内的位置处的视觉变化，用于来自图像集合的图像中描绘的对象的给定物理特征 与图像相关联。 第一和第二3D几何图形描绘了不同时间位置处物体的物理特征。 可以基于每组的确定的视觉变化对图像集进行排名。 可以选择来自该组图像的一组给定的图像，以便基于排名在用户界面上显示。

公开(公告)号：US20160027177A1

公开(公告)日：2016-01-28

申请号：US14337680

申请日：2014-07-22

Applicant: Google Inc.

Inventor： Allen Hutchison

IPC: G06T7/00 ,  G06K9/62

CPC classification number: G06K9/6202 ,  G06K9/00664 ,  G06K9/6267 ,  H04N5/23229

Abstract: Systems and methods are provided for using imagery depicting a timekeeping device to determine a clock offset for a particular image capture device. The clock offset can be used to correct timestamps associated with one or more images captured by such image capture device. One example method includes analyzing imagery depicting at least in part a timekeeping device to determine a first time displayed by the timekeeping device in the imagery. The method includes determining whether the first time comprises a 12-hour value or a 24-hour value. The method includes, when it is determined that the first time comprises a 12-hour value, determining a corresponding 24-hour value for the 12-hour value based at least in part on information contained within a plurality of images. The method includes determining a clock offset between the 24-hour value and the first timestamp. One example system includes a timestamp correction engine for correcting timestamps.

Abstract translation: 提供的系统和方法用于使用描绘计时装置的图像来确定特定图像捕获装置的时钟偏移。 时钟偏移可用于校正与由这种图像捕获设备捕获的一个或多个图像相关联的时间戳。 一个示例性方法包括分析至少部分地描绘计时装置以确定由图像中的计时装置显示的第一次的图像。 该方法包括确定第一次是否包括12小时值或24小时值。 该方法包括：当确定第一次包括12小时值时，至少部分地基于包含在多个图像中的信息来确定对于12小时值的相应24小时值。 该方法包括确定24小时值和第一时间戳之间的时钟偏移。 一个示例系统包括用于校正时间戳的时间戳校正引擎。