公开(公告)号：US10068352B2

公开(公告)日：2018-09-04

申请号：US15279103

申请日：2016-09-28

Applicant: Google Inc.

Inventor： Sundeep Singh Khandpur

IPC: G06T7/40 ,  G06T7/00 ,  G06T11/00 ,  G09G5/02 ,  G09G5/393 ,  H04N1/60 ,  G06T7/90 ,  G06Q30/02 ,  G06T7/33 ,  G06K9/46

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining whether content rendered on a display is actually visible to a user. In one aspect, a method includes generating and transmitting content presentation data that causes a user device to present, at a display and over a first time period, a content item that includes one or more content portions that are each designated for presentation in a respective first color. For a second time period, the respective first color of a particular content portion is changed to a second color. Visual representation data that includes a visual representation of the display during the second time period is generated and provided. Using the visual representation, an amount of the content item that was visible at the display is determined based on an amount of the particular content portion presented in the second color.

公开(公告)号：US20150163448A1

公开(公告)日：2015-06-11

申请号：US14099781

申请日：2013-12-06

Applicant: Google Inc.

Inventor： Sundeep Singh Khandpur ,  John Mathias Barr

IPC: H04N5/77 ,  G06K9/00

CPC classification number: G06K9/00221 ,  G06K9/00979 ,  G06K2009/00328 ,  H04N5/772

Abstract: A system and machine-implemented method for real-time facial recognition at an image capture device is provided. The method includes capturing an image at an image capture device, the image containing at least a first person, analyzing the image to identify one or more data points corresponding to the first person, determining if mapping information corresponding to the one or more data points is locally available at the image capture device, sending a request to a remote device for mapping information when it is determined that mapping information corresponding to the one or more data points is not locally available at the image capture device, receiving mapping information corresponding to the one or more data points in response to the request and locally storing the mapping information at the image capture device.

Abstract translation: 提供了一种用于图像捕获设备实时面部识别的系统和机器实现方法。 该方法包括在图像捕获装置处拍摄图像，该图像至少包含第一人物，分析图像以识别与第一人对应的一个或多个数据点，确定与该一个或多个数据点相对应的映射信息是否是 在图像捕获设备本地可用的情况下，当确定与所述一个或多个数据点相对应的映射信息在所述图像捕获设备处不可本地可用时，向远程设备发送请求以映射信息;接收对应于所述一个或多个数据点的映射信息 或多个数据点，并且在图像捕获装置处本地存储映射信息。

公开(公告)号：US20180012378A1

公开(公告)日：2018-01-11

申请号：US15279103

申请日：2016-09-28

Applicant: Google Inc.

Inventor： Sundeep Singh Khandpur

IPC: G06T11/00 ,  G06Q30/02 ,  G06K9/46

CPC classification number: G06T7/90 ,  G06K9/4652 ,  G06Q30/0242 ,  G06Q30/0277 ,  G06T7/33 ,  G06T7/62 ,  G06T11/001 ,  G06T2207/10004 ,  G06T2207/10024 ,  G06T2207/30242

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining whether content rendered on a display is actually visible to a user. In one aspect, a method includes generating and transmitting content presentation data that causes a user device to present, at a display and over a first time period, a content item that includes one or more content portions that are each designated for presentation in a respective first color. For a second time period, the respective first color of a particular content portion is changed to a second color. Visual representation data that includes a visual representation of the display during the second time period is generated and provided. Using the visual representation, an amount of the content item that was visible at the display is determined based on an amount of the particular content portion presented in the second color.

公开(公告)号：US09569656B2

公开(公告)日：2017-02-14

申请号：US14099781

申请日：2013-12-06

Applicant: Google Inc.

Inventor： Sundeep Singh Khandpur ,  John Mathias Barr

IPC: H04N9/47 ,  H04N7/18 ,  G06K9/00 ,  H04N5/77

CPC classification number: G06K9/00221 ,  G06K9/00979 ,  G06K2009/00328 ,  H04N5/772

Abstract: A system and machine-implemented method for real-time facial recognition at an image capture device is provided. The method includes capturing an image at an image capture device, the image containing at least a first person, analyzing the image to identify one or more data points corresponding to the first person, determining if mapping information corresponding to the one or more data points is locally available at the image capture device, sending a request to a remote device for mapping information when it is determined that mapping information corresponding to the one or more data points is not locally available at the image capture device, receiving mapping information corresponding to the one or more data points in response to the request and locally storing the mapping information at the image capture device.

Abstract translation: 提供了一种用于图像捕获设备实时面部识别的系统和机器实现方法。 该方法包括在图像捕获装置处拍摄图像，该图像至少包含第一人物，分析图像以识别与第一人对应的一个或多个数据点，确定与该一个或多个数据点相对应的映射信息是否是 在图像捕获设备本地可用的情况下，当确定与所述一个或多个数据点相对应的映射信息在所述图像捕获设备处不可本地可用时，向远程设备发送请求以映射信息;接收对应于所述一个或多个数据点的映射信息 或多个数据点，并且在图像捕获装置处本地存储映射信息。

公开(公告)号：US09478042B1

公开(公告)日：2016-10-25

申请号：US14458082

申请日：2014-08-12

Applicant: Google Inc.

Inventor： Sundeep Singh Khandpur

IPC: G06T7/40 ,  G06Q30/02 ,  G09G5/02 ,  G09G5/39 ,  G06F3/0481 ,  H04N1/60 ,  G06T11/00

CPC classification number: G06T7/90 ,  G06K9/4652 ,  G06Q30/0242 ,  G06Q30/0277 ,  G06T7/33 ,  G06T7/62 ,  G06T11/001 ,  G06T2207/10004 ,  G06T2207/10024 ,  G06T2207/30242

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining whether content rendered on a display is actually visible to a user. In one aspect, a method includes generating and transmitting content presentation data that causes a user device to present, at a display and over a first time period, a content item that includes one or more content portions that are each designated for presentation in a respective first color. For a second time period, the respective first color of a particular content portion is changed to a second color. Visual representation data that includes a visual representation of the display during the second time period is generated and provided. Using the visual representation, an amount of the content item that was visible at the display is determined based on an amount of the particular content portion presented in the second color.

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，用于确定呈现在显示器上的内容是否实际上对用户可见。 一方面，一种方法包括生成和发送内容呈现数据，该内容呈现数据使得用户设备在显示器和第一时间周期内呈现包括一个或多个内容部分的内容项目，每个内容部分被指定用于呈现在相应的 第一种颜色 对于第二时间段，特定内容部分的相应第一颜色被改变为第二颜色。 生成并提供在第二时间段期间包括显示器的视觉表示的视觉表示数据。 使用视觉表示，基于以第二颜色呈现的特定内容部分的量来确定在显示器处可见的内容项目的量。