公开(公告)号：US20110295469A1

公开(公告)日：2011-12-01

申请号：US12008430

申请日：2008-01-11

申请人： Abbas Rafii ,  Richard New ,  Sunil Acharya ,  Timothy Droz

发明人： Abbas Rafii ,  Richard New ,  Sunil Acharya ,  Timothy Droz

IPC分类号： E05F15/02 ,  G06F17/00

CPC分类号： E05F15/43 ,  E05F2015/434 ,  E05F2015/483 ,  E05Y2900/531 ,  G06T7/254 ,  G06T2207/10028 ,  G06T2207/30248

摘要： Time-of-flight (TOF) three-dimensional sensing systems are deployed on or in a motor vehicle to image contact zones associated with potential contact between an avoidable object and the vehicle or vehicle frame and/or remotely controllable motorized moving door or liftgate. An algorithm processes depth data acquired by each TOF system to determine whether an avoidable object is in the associated contact zone. If present, a control signal issues to halt or reverse the mechanism moving the door. A stored database preferably includes a depth image of the contact zone absent any object, an image of the door, and volume of the door. Database images are compared to newly acquired depth images to identify pixel sensors whose depth values are statistically unlikely to represent background or the door. Pixels within the contact zone so identified are an object, and the control signal is issued.

摘要翻译： 飞行时间（TOF）三维感测系统部署在机动车辆上或机动车辆中，以对与可避免物体与车辆或车辆框架和/或远程可控的电动移动门或升降门之间的潜在接触相关联的图像接触区进行成像。 一种算法处理由每个TOF系统获取的深度数据，以确定一个可避免的对象是否在相关联的接触区域中。 如果存在，则控制信号出现停止或反转移动门的机构。 存储的数据库优选地包括没有任何对象的接触区域的深度图像，门的图像和门的体积。 将数据库图像与新获取的深度图像进行比较，以识别其深度值在统计上不太可能代表背景或门的像素传感器。 所识别的接触区域内的像素是对象，并且发出控制信号。

公开(公告)号：US08589033B2

公开(公告)日：2013-11-19

申请号：US12008430

申请日：2008-01-11

申请人： Abbas Rafii ,  Richard New ,  Sunil Acharya ,  Timothy Droz

发明人： Abbas Rafii ,  Richard New ,  Sunil Acharya ,  Timothy Droz

IPC分类号： E05F15/20 ,  G01C3/00 ,  G05B19/4061 ,  B60J5/00

CPC分类号： E05F15/43 ,  E05F2015/434 ,  E05F2015/483 ,  E05Y2900/531 ,  G06T7/254 ,  G06T2207/10028 ,  G06T2207/30248

摘要： Time-of-flight (TOF) three-dimensional sensing systems are deployed on or in a motor vehicle to image contact zones associated with potential contact between an avoidable object and the vehicle or vehicle frame and/or remotely controllable motorized moving door or liftgate. An algorithm processes depth data acquired by each TOF system to determine whether an avoidable object is in the associated contact zone. If present, a control signal issues to halt or reverse the mechanism moving the door. A stored database preferably includes a depth image of the contact zone absent any object, an image of the door, and volume of the door. Database images are compared to newly acquired depth images to identify pixel sensors whose depth values are statistically unlikely to represent background or the door. Pixels within the contact zone so identified are an object, and the control signal is issued.

摘要翻译： 飞行时间（TOF）三维感测系统部署在机动车辆上或机动车辆中，以对与可避免物体与车辆或车辆框架和/或远程可控的电动移动门或升降门之间的潜在接触相关联的图像接触区进行成像。 一种算法处理由每个TOF系统获取的深度数据，以确定一个可避免的对象是否在相关联的接触区域中。 如果存在，则控制信号出现停止或反转移动门的机构。 存储的数据库优选地包括没有任何对象的接触区域的深度图像，门的图像和门的体积。 将数据库图像与新获取的深度图像进行比较，以识别其深度值在统计上不太可能代表背景或门的像素传感器。 所识别的接触区域内的像素是对象，并且发出控制信号。

公开(公告)号：US20120011454A1

公开(公告)日：2012-01-12

申请号：US12387438

申请日：2009-04-30

申请人： Timothy Droz ,  Sunil Acharya ,  Cyrus Bamji

发明人： Timothy Droz ,  Sunil Acharya ,  Cyrus Bamji

IPC分类号： G06F3/00

CPC分类号： H04L12/1827 ,  G06Q30/0241 ,  G06Q30/0251

摘要： The present invention mines or extracts data present during interaction between at least two participants, for example in a chat session, a video session, etc. via the Internet. The data, which can include participant web camera generated video, audio, keyboard typed information, handwriting recognized information, is analyzed. Based upon the analysis, content-dependent information is determined and may be displayed to one or more participants in the chat session. In one aspect, a video foreground based upon a participant's generated video is combined with a customized computer generated background that is based upon data mined from the chat session. The customized background preferably is melded seamlessly with the participant's foreground data, preferably via background substitution that combines RGB video with depth data that predicts what background may substituted with new imagery. Content-based targeted information can include advertisement(s).

摘要翻译： 本发明例如在通过因特网的聊天会话，视频会话等中，在至少两个参与者之间的交互期间挖掘或提取存在的数据。 分析了可以包括参与者网络摄像机产生的视频，音频，键盘类型信息，手写识别信息的数据。 基于分析，确定内容相关信息并且可以向聊天会话中的一个或多个参与者显示。 在一个方面，基于参与者生成的视频的视频前景与基于从聊天会话中挖掘的数据的定制的计算机生成的背景组合。 优选地，定制背景与参与者的前景数据无缝地融合，优选地通过将RGB视频与预测可以用新图像替代的背景的深度数据组合的背景替换。 基于内容的目标信息可以包括广告。