专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

11. 发明授权

US10592611B2 System for automatic extraction of structure from spoken conversation using lexical and acoustic features 有权
公开(公告)号：US10592611B2
公开(公告)日：2020-03-17
申请号：US15332766
申请日：2016-10-24
申请人： Palo Alto Research Center Incorporated
发明人： Jesse Vig , Harish Arsikere , Margaret H. Szymanski , Luke R. Plurkowski , Kyle D. Dent , Daniel G. Bobrow , Daniel Davies , Eric Saund
IPC分类号： G10L15/00 , G06F17/27 , G10L25/48 , G10L15/26 , H04M3/51
摘要： Embodiments of the present invention provide a system for automatically extracting conversational structure from a voice record based on lexical and acoustic features. The system also aggregates business-relevant statistics and entities from a collection of spoken conversations. The system may infer a coarse-level conversational structure based on fine-level activities identified from extracted acoustic features. The system improves significantly over previous systems by extracting structure based on lexical and acoustic features. This enables extracting conversational structure on a larger scale and finer level of detail than previous systems, and can feed an analytics and business intelligence platform, e.g. for customer service phone calls. During operation, the system obtains a voice record. The system then extracts a lexical feature using automatic speech recognition (ASR). The system extracts an acoustic feature. The system then determines, via machine learning and based on the extracted lexical and acoustic features, a coarse-level structure of the conversation.

12. 发明授权

US10296570B2 Reflow narrative text objects in a document having text objects and graphical objects, wherein text object are classified as either narrative text object or annotative text object based on the distance from a left edge of a canvas of display 有权
公开(公告)号：US10296570B2
公开(公告)日：2019-05-21
申请号：US14062971
申请日：2013-10-25
申请人： Palo Alto Research Center Incorporated
发明人： Eric Saund , James V. Mahoney , William C. Janssen, Jr.
IPC分类号： G06F17/21 , G06F17/22 , G06F17/24
摘要： A method and system is provided to author an electronic document having content including document objects, the documents objects including at least text objects and graphical objects, the graphical objects including at least one of image objects, structured graphics, digital ink, and hyperlinks. The method and system modifies the content of the electronic document, and applies an automatic hybrid reflow process to reflow at least one of the text objects of the electronic document in accordance with the applied automatic hybrid reflow process. The applied automatic hybrid reflow process acting to position the text objects and the graphical objects on the electronic document.

13. 发明申请

US20140334668A1 SYSTEM AND METHOD FOR VISUAL MOTION BASED OBJECT SEGMENTATION AND TRACKING 有权
标题翻译：基于视觉运动的对象分类与跟踪系统与方法
公开(公告)号：US20140334668A1
公开(公告)日：2014-11-13
申请号：US13891653
申请日：2013-05-10
申请人： PALO ALTO RESEARCH CENTER INCORPORATED
发明人： Eric Saund
IPC分类号： G06T7/20
CPC分类号： G06T7/20 , B64C39/024 , B64C2201/127 , B64D47/08 , G06T7/215 , G06T7/246 , G06T2207/10032 , G06T2207/30236 , G06T2207/30252 , G06T2207/30256
摘要： The PMP Growth algorithm described herein provides for image tracking, segmentation and processing in environments where the camera system moves around a great deal, i.e., causing image jumps from one image frame to the next. It also is operative in systems where the objects themselves are making quick movements that alter their path. Attributes of the PMP Growth algorithm allow tracking systems using the PMP Growth algorithm to follow objects a long distance in a scene. This detection and tracking method is designed to track objects within a sequence of video image frames, and includes detecting keypoints in a current image frame of the video image frames, assigning local appearance features to the detected keypoints, establishing Point-Motion-Pairs between two successive image frames of the video image frames, and accumulating additional matches between image locations to form complete coherent motion object models of the objects being tracked. The segmentation aspect permits for the discovery of different coherently moving regions in the images.
摘要翻译：本文描述的PMP生长算法提供了在相机系统大量移动的环境中的图像跟踪，分割和处理，即使图像从一个图像帧跳到下一个图像帧。它也可以在物体本身正在快速移动以改变其路径的系统中运行。 PMP成长算法的属性允许使用PMP Growth算法的跟踪系统在场景中跟踪对象长距离。该检测和跟踪方法被设计为跟踪视频图像帧序列内的对象，并且包括检测视频图像帧的当前图像帧中的关键点，将检测到的关键点分配局部外观特征，在两个之间建立点运动对视频图像帧的连续图像帧，以及在图像位置之间累积附加匹配，以形成被跟踪对象的完整的相干运动对象模型。分割方面允许在图像中发现不同的相干移动区域。

14. 发明申请

US20130257874A1 METHOD FOR GENERATING A GRAPH LATTICE FROM A CORPUS OF ONE OR MORE DATA GRAPHS 有权
标题翻译：从一个或多个数据图形的公司生成图形格式的方法
公开(公告)号：US20130257874A1
公开(公告)日：2013-10-03
申请号：US13906757
申请日：2013-05-31
申请人： Palo Alto Research Center Incorporated
发明人： Eric Saund
IPC分类号： G06T11/20
CPC分类号： G06T11/206
摘要： A document recognition system and method, where images are represented as a collection of primitive features whose spatial relations are represented as a graph. Useful subsets of all the possible subgraphs representing different portions of images are represented over a corpus of many images. The data structure is a lattice of subgraphs, and algorithms are provided means to build and use the graph lattice efficiently and effectively.
摘要翻译：一种文档识别系统和方法，其中图像被表示为其空间关系被表示为图形的原始特征的集合。表示图像的不同部分的所有可能子图的有用子集在许多图像的语料库上表示。数据结构是子图的格子，提供了有效和高效地构建和使用图形格子的算法。

15. 发明授权

US09025825B2 System and method for visual motion based object segmentation and tracking 有权
标题翻译：基于视觉运动的物体分割和跟踪的系统和方法
公开(公告)号：US09025825B2
公开(公告)日：2015-05-05
申请号：US13891653
申请日：2013-05-10
申请人： Palo Alto Research Center Incorporated
发明人： Eric Saund
IPC分类号： G06K9/00 , G06T7/20
CPC分类号： G06T7/20 , B64C39/024 , B64C2201/127 , B64D47/08 , G06T7/215 , G06T7/246 , G06T2207/10032 , G06T2207/30236 , G06T2207/30252 , G06T2207/30256
摘要： The PMP Growth algorithm described herein provides for image tracking, segmentation and processing in environments where the camera system moves around a great deal, i.e., causing image jumps from one image frame to the next. It also is operative in systems where the objects themselves are making quick movements that alter their path. Attributes of the PMP Growth algorithm allow tracking systems using the PMP Growth algorithm to follow objects a long distance in a scene. This detection and tracking method is designed to track objects within a sequence of video image frames, and includes detecting keypoints in a current image frame of the video image frames, assigning local appearance features to the detected keypoints, establishing Point-Motion-Pairs between two successive image frames of the video image frames, and accumulating additional matches between image locations to form complete coherent motion object models of the objects being tracked. The segmentation aspect permits for the discovery of different coherently moving regions in the images.
摘要翻译：本文描述的PMP生长算法提供了在相机系统大量移动的环境中的图像跟踪，分割和处理，即使图像从一个图像帧跳到下一个图像帧。它也可以在物体本身正在快速移动以改变其路径的系统中运行。 PMP成长算法的属性允许使用PMP Growth算法的跟踪系统在场景中跟踪对象长距离。该检测和跟踪方法被设计为跟踪视频图像帧序列内的对象，并且包括检测视频图像帧的当前图像帧中的关键点，将检测到的关键点分配局部外观特征，在两个之间建立点运动对视频图像帧的连续图像帧，以及在图像位置之间累积附加匹配，以形成被跟踪对象的完整的相干运动对象模型。分割方面允许在图像中发现不同的相干移动区域。

16. 发明申请

US20150121305A1 METHOD AND SYSTEM FOR ENHANCED INFERRED MODE USER INTERFACE OPERATIONS 有权
标题翻译：用于增强模式用户界面操作的方法和系统
公开(公告)号：US20150121305A1
公开(公告)日：2015-04-30
申请号：US14063002
申请日：2013-10-25
申请人： Palo Alto Research Center Incorporated
发明人： Eric Saund , James V. Mahoney , William C. Janssen, JR.
IPC分类号： G06F3/0484 , G06F3/0488
CPC分类号： G06Q10/107
摘要： A method and system is provided for the input of user interface commands. Particularly, command initiation events including at least one of: (i) a mouse press event on mouse pointer hardware, and (ii) at least one of a pen touch event, a stylus touch event, and a finger touch event on touch pointer hardware, are accepted. Then gesture stroke input events including at least one of: (i) a mouse drag event on the mouse pointer hardware, and (ii) at least one of a pen drag event, a stylus drag event and a finger drag event on the touch pointer hardware, are accepted. Additionally, command termination events including at least one of (i) a mouse release event on the mouse pointer hardware, and (ii) at least one of a pen lift event, a stylus lift event and a finger lift event on the touch pointer hardware, are accepted. The events are then interpreted as at least one of object selection or digital ink input operations without prior selection of a user input mode.
摘要翻译：提供了一种用于输入用户界面命令的方法和系统。特别地，命令启动事件包括以下中的至少一个：（i）鼠标指针硬件上的鼠标按压事件，以及（ii）触摸指针硬件上的笔触摸事件，触笔触摸事件和手指触摸事件中的至少一个，被接受。然后，手势笔画输入事件包括以下中的至少一个：（i）鼠标指针硬件上的鼠标拖动事件，以及（ii）触摸指示器上的笔拖动事件，触笔拖动事件和手指拖动事件中的至少一个硬件，被接受。另外，指令终止事件包括（i）鼠标指针硬件上的鼠标释放事件中的至少一个，以及（ii）触摸指针硬件上的笔提示事件，触控笔提升事件和手指提升事件中的至少一个，被接受。然后将事件解释为目标选择或数字墨水输入操作中的至少一个，而无需事先选择用户输入模式。

17. 发明授权

US08972425B2 Efficient globally optimal interpretation of documents 有权
标题翻译：有效的全球最佳解读文件
公开(公告)号：US08972425B2
公开(公告)日：2015-03-03
申请号：US13799568
申请日：2013-03-13
申请人： Palo Alto Research Center Incorporated
发明人： Christina Pavlopoulou , Evgeniy Bart , Eric Saund
IPC分类号： G06F17/30
CPC分类号： G06F17/2715 , G06F17/2294
摘要： A method is provided for parsing a document having a plurality of lines on which items are listed spanning one or more lines. It includes: obtaining a plurality of candidates, representing hypothetical items within the document, each candidate spanning one or more lines and having a local cost representing a confidence in a quality of the candidate compared to a model; determining labeling costs for intervals of the document defined between pairs of lines, each interval containing candidates therein, each labeling cost reflecting a configuration of the candidates within the interval; identifying a best labeling for each interval based on the labeling costs determined for that interval, the best labeling corresponding to one of the configurations of the candidates within the interval; defining a global objective function; and selecting a subset of the candidates such that the global objective function is optimized, based on the identified best labelings.
摘要翻译：提供了一种用于解析具有多个行的文档的方法，其中列出了跨越一行或多行的项目。它包括：获得多个候选人，代表文档内的假想项目，跨越一行或多行的每个候选人并且具有表示与模型相比候选者的质量的置信度的本地成本; 确定在线对之间定义的文档的间隔的标签成本，每个间隔包含其中的候选，每个标签成本反映该间隔内候选的配置; 基于为该间隔确定的标签成本，确定每个间隔的最佳标签，对应于间隔内的候选者的配置之一的最佳标签; 定义全局目标函数; 以及基于所识别的最佳标签来选择所述候选的子集，使得全局目标函数被优化。

18. 发明申请

US20140336848A1 SYSTEM AND METHOD FOR DETECTING, TRACKING AND ESTIMATING THE SPEED OF VEHICLES FROM A MOBILE PLATFORM 有权
标题翻译：用于从移动平台检测，跟踪和估计车辆速度的系统和方法
公开(公告)号：US20140336848A1
公开(公告)日：2014-11-13
申请号：US13891765
申请日：2013-05-10
申请人： Palo Alto Research Center Incorporated
发明人： Eric Saund , Christopher Paulson , Gregory Burton , Eric Peeters
IPC分类号： G08G1/054
CPC分类号： G08G1/054 , B64C2201/127 , G05D1/0094 , G06K9/0063 , G06K9/00785 , G06K9/3241 , G06K2209/23 , G06T7/215 , G06T7/246 , G06T2207/10032 , G06T2207/30236 , G08G1/0175 , G08G5/00
摘要： A method and system for measurement of ground based vehicle speed includes a movable platform that includes an unmanned aerial vehicle (UAV) located in proximity to a roadway, the UAV operates under control and navigation of a UAV control unit, and the UAV also carries camera and monitoring equipment, the camera and monitoring equipment including an onboard computing system, and a camera with a wide angle lens and a camera with a telephoto lens, the cameras being mounted on a pan/tilt device. An algorithm operated by the on-board computing system is used to detect and track vehicles moving on a roadway. The algorithm is configured to detect and track the vehicles despite motion created by movement of the UAV. The cameras mounted on the pan/tilt device are moved under the direction of the computer vision algorithm to maintain a target vehicle of the detected moving vehicles in view, and the speed of the target vehicle is measured.
摘要翻译：一种用于测量地面车辆速度的方法和系统包括可移动平台，其包括位于巷道附近的无人驾驶飞行器（UAV），UAV在UAV控制单元的控制和导航下操作，并且UAV还携带摄像机以及监视设备，摄像机和包括机载计算系统的监视设备，以及具有广角镜头的照相机和具有长焦镜头的照相机，照相机安装在平移/倾斜装置上。使用由车载计算系统操作的算法来检测和跟踪在道路上行驶的车辆。该算法被配置为检测和跟踪车辆，尽管由UAV的移动产生的运动。安装在摇摄/倾斜装置上的摄像机在计算机视觉算法的方向下移动以保持检测到的移动车辆的目标车辆的视野，并且测量目标车辆的速度。

19. 发明申请

US20140280078A1 EFFICIENT GLOBALLY OPTIMAL INTERPRETATION OF DOCUMENTS 有权
标题翻译：有效的全面的文件解释
公开(公告)号：US20140280078A1
公开(公告)日：2014-09-18
申请号：US13799568
申请日：2013-03-13
申请人： PALO ALTO RESEARCH CENTER INCORPORATED
发明人： Christina Pavlopoulou , Evgeniy Bart , Eric Saund
IPC分类号： G06F17/30
CPC分类号： G06F17/2715 , G06F17/2294
摘要： A method is provided for parsing a document having a plurality of lines on which items are listed spanning one or more lines. It includes: obtaining a plurality of candidates, representing hypothetical items within the document, each candidate spanning one or more lines and having a local cost representing a confidence in a quality of the candidate compared to a model; determining labeling costs for intervals of the document defined between pairs of lines, each interval containing candidates therein, each labeling cost reflecting a configuration of the candidates within the interval; identifying a best labeling for each interval based on the labeling costs determined for that interval, the best labeling corresponding to one of the configurations of the candidates within the interval; defining a global objective function; and selecting a subset of the candidates such that the global objective function is optimized, based on the identified best labelings.
摘要翻译：提供了一种用于解析具有多个行的文档的方法，其中列出了跨越一行或多行的项目。它包括：获得多个候选人，代表文档内的假想项目，跨越一行或多行的每个候选人，并且具有表示候选人与模型相比质量的置信度的本地成本; 确定在线对之间定义的文档的间隔的标签成本，每个间隔包含其中的候选，每个标签成本反映该间隔内候选的配置; 基于为该间隔确定的标签成本，确定每个间隔的最佳标签，对应于间隔内的候选者的配置之一的最佳标签; 定义全局目标函数; 以及基于所识别的最佳标签来选择所述候选的子集，使得全局目标函数被优化。

20. 发明申请

US20130114890A1 SYSTEM AND METHOD FOR SEGMENTING TEXT LINES IN DOCUMENTS 有权
标题翻译：用于在文档中分隔文本行的系统和方法
公开(公告)号：US20130114890A1
公开(公告)日：2013-05-09
申请号：US13677473
申请日：2012-11-15
申请人： Palo Alto Research Center Incorporated
发明人： Eric Saund
IPC分类号： G06K9/00
CPC分类号： G06K9/00442 , G06K9/00449 , G06K9/00456 , G06K9/346 , G06K2209/01
摘要： Methods and systems of the present embodiment provide segmenting of connected components of markings found in document images. Segmenting includes detecting aligned text. From this detected material an aligned text mask is generated and used in processing of the images. The processing includes breaking connected components in the document images into smaller pieces or fragments by detecting and segregating the connected components and fragments thereof likely to belong to aligned text.
摘要翻译：本实施例的方法和系统提供在文档图像中找到的标记的连接分量的分割。分段包括检测对齐的文本。从该检测到的材料中，生成对齐的文本掩码并用于图像的处理。该处理包括通过检测和分离可能属于对齐文本的连接的组件及其片段来将文档图像中的连接的组件分解成更小的片段或片段。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式