专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20130343642A1 MACHINE-LEARNT PERSON RE-IDENTIFICATION 有权
标题翻译：机器学习人员重新认证
公开(公告)号：US20130343642A1
公开(公告)日：2013-12-26
申请号：US13913685
申请日：2013-06-10
申请人： Cheng-Hao Kuo , Vinay Damodar Shet , Sameh Khamis , Larry Davis , Vivek Kumar Singh
发明人： Cheng-Hao Kuo , Vinay Damodar Shet , Sameh Khamis , Larry Davis , Vivek Kumar Singh
IPC分类号： G06K9/62
CPC分类号： G06K9/6256 , G06K9/4652
摘要： Automated person re-identification may be assisted by consideration of attributes of the person in a joint classification with matching of the person. By both solving for similarities in a plurality of attributes and identities, discriminative interactions may be captured. Automated person re-identification may be assisted by consideration of a semantic color name. Rather than a color histogram, probability distributions are mapped to color terms of the semantic color name. Using other descriptors as well, similarity measures for the various descriptors are weighted and combined into a score. Either or both considerations may be used.
摘要翻译：自动人员重新识别可以通过考虑与人的匹配的联合分类中的人的属性来辅助。通过解决多个属性和身份中的相似性，可以捕获歧视性交互。可以通过考虑语义颜色名称来辅助自动人员重新识别。概率分布不是颜色直方图，而是映射到语义颜色名称的颜色项。还使用其他描述符，对各种描述符的相似性度量进行加权并组合成一个分数。可以使用这两种或两种考虑因素。

2. 发明授权

US09396412B2 Machine-learnt person re-identification 有权
标题翻译：机器人重新识别
公开(公告)号：US09396412B2
公开(公告)日：2016-07-19
申请号：US13913685
申请日：2013-06-10
申请人： Cheng-Hao Kuo , Vinay Damodar Shet , Sameh Khamis
发明人： Cheng-Hao Kuo , Vinay Damodar Shet , Sameh Khamis
IPC分类号： G06K9/62 , G06K9/46
CPC分类号： G06K9/6256 , G06K9/4652
摘要： Automated person re-identification may be assisted by consideration of attributes of the person in a joint classification with matching of the person. By both solving for similarities in a plurality of attributes and identities, discriminative interactions may be captured. Automated person re-identification may be assisted by consideration of a semantic color name. Rather than a color histogram, probability distributions are mapped to color terms of the semantic color name. Using other descriptors as well, similarity measures for the various descriptors are weighted and combined into a score. Either or both considerations may be used.
摘要翻译：自动人员重新识别可以通过考虑与人的匹配的联合分类中的人的属性来辅助。通过解决多个属性和身份中的相似性，可以捕获歧视性交互。可以通过考虑语义颜色名称来辅助自动人员重新识别。概率分布不是颜色直方图，而是映射到语义颜色名称的颜色项。还使用其他描述符，对各种描述符的相似性度量进行加权并组合成一个分数。可以使用这两种或两种考虑因素。

3. 发明授权

US09117147B2 Marginal space learning for multi-person tracking over mega pixel imagery 有权
标题翻译：用于多人跟踪超大像素图像的边缘空间学习
公开(公告)号：US09117147B2
公开(公告)日：2015-08-25
申请号：US13451845
申请日：2012-04-20
申请人： Vinay Damodar Shet , Dorin Comaniciu , Sushil Mittal , Peter Meer , Cheng-Hao Kuo
发明人： Vinay Damodar Shet , Dorin Comaniciu , Sushil Mittal , Peter Meer , Cheng-Hao Kuo
IPC分类号： H04N5/225 , G06K9/68 , G06T7/20 , G08B13/196
CPC分类号： G06K9/6857 , G06T7/277 , G06T2207/20081 , G06T2207/30196 , G06T2207/30232 , G06T2207/30241 , G08B13/19608
摘要： A method for tracking pedestrians in a video sequence, where each image frame of the video sequence corresponds to a time step, includes using marginal space learning to sample a prior probability distribution p(xt|Zt−1) of multi-person identity assignments given a set of feature measurements from all previous image frames, using marginal space learning to estimate an observation likelihood distribution p(zt|xt) of the set of features given a set of multi-person identity assignments sampled from the prior probability distribution, calculating a posterior probability distribution p(xt|Zt) from the observation likelihood distribution p(zt|xt) and the prior probability distribution p(xt|Zt−1), and using marginal space learning to estimate the prior probability distribution p(xt+1|Zt) for a next image frame given the posterior probability distribution p(xt|Zt) and a probability p(xt+1|xt), where the posterior probability distribution of multi-person identity assignments corresponds to a set of pedestrian detection hypotheses for the video sequence.
摘要翻译：一种用于跟踪视频序列中的行人的方法，其中视频序列的每个图像帧对应于时间步长，包括使用边缘空间学习来采样给定的多人身份分配的先验概率分布p（xt | Zt-1）一组来自所有先前图像帧的特征测量，使用边缘空间学习来估计给定从先验概率分布采样的一组多人身份分配的特征集合的观察似然分布p（zt | xt），计算根据观察似然分布p（zt | xt）和先验概率分布p（xt | Zt-1）的后验概率分布p（xt | Zt），并利用边际空间学习估计先验概率分布p（xt + | Zt），给出后验概率分布p（xt | Zt）和概率p（xt + 1 | xt）的下一图像帧，其中多人身份分配的后验概率分布对应于视频序列的一组行人检测假设。

4. 发明申请

US20120274781A1 MARGINAL SPACE LEARNING FOR MULTI-PERSON TRACKING OVER MEGA PIXEL IMAGERY 有权
标题翻译：用于MEGA像素图像的多人跟踪的边缘空间学习
公开(公告)号：US20120274781A1
公开(公告)日：2012-11-01
申请号：US13451845
申请日：2012-04-20
申请人： Vinay Damodar Shet , Dorin Comaniciu , Sushil Mittal , Peter Meer , Cheng-Hao Kuo
发明人： Vinay Damodar Shet , Dorin Comaniciu , Sushil Mittal , Peter Meer , Cheng-Hao Kuo
IPC分类号： G06K9/62 , H04N5/225
CPC分类号： G06K9/6857 , G06T7/277 , G06T2207/20081 , G06T2207/30196 , G06T2207/30232 , G06T2207/30241 , G08B13/19608
摘要： A method for tracking pedestrians in a video sequence, where each image frame of the video sequence corresponds to a time step, includes using marginal space learning to sample a prior probability distribution p(xt|Zt−1) of multi-person identity assignments given a set of feature measurements from all previous image frames, using marginal space learning to estimate an observation likelihood distribution p(zt|xt) of the set of features given a set of multi-person identity assignments sampled from the prior probability distribution, calculating a posterior probability distribution p(xt|Zt) from the observation likelihood distribution p(zt|xt) and the prior probability distribution p(xt|Zt−1), and using marginal space learning to estimate the prior probability distribution p(xt+1|Zt) for a next image frame given the posterior probability distribution p(xt|Zt) and a probability p(xt+1|xt), where the posterior probability distribution of multi-person identity assignments corresponds to a set of pedestrian detection hypotheses for the video sequence.
摘要翻译：一种用于跟踪视频序列中的行人的方法，其中视频序列的每个图像帧对应于时间步长，包括使用边缘空间学习来采样给定的多人身份分配的先验概率分布p（xt | Zt-1）一组来自所有先前图像帧的特征测量，使用边缘空间学习来估计给定从先验概率分布采样的一组多人身份分配的特征集合的观察似然分布p（zt | xt），计算根据观察似然分布p（zt | xt）和先验概率分布p（xt | Zt-1）的后验概率分布p（xt | Zt），并利用边际空间学习估计先验概率分布p（xt + | Zt），给出后验概率分布p（xt | Zt）和概率p（xt + 1 | xt）的下一图像帧，其中多人身份分配的后验概率分布对应于视频序列的一组行人检测假设。

5. 发明申请

US20120243741A1 Object Recognition For Security Screening and Long Range Video Surveillance 有权
标题翻译：对象识别用于安全检查和远程视频监控
公开(公告)号：US20120243741A1
公开(公告)日：2012-09-27
申请号：US13397924
申请日：2012-02-16
申请人： Vinay Damodar Shet , Claus Bahlmann , Maneesh Kumar Singh
发明人： Vinay Damodar Shet , Claus Bahlmann , Maneesh Kumar Singh
IPC分类号： G06K9/62 , G06K9/00
CPC分类号： G06K9/6296 , G06K2209/09
摘要： A method of detecting an object in image data that is deemed to be a threat includes annotating sections of at least one training image to indicate whether each section is a component of the object, encoding a pattern grammar describing the object using a plurality of first order logic based predicate rules, training distinct component detectors to each identify a corresponding one of the components based on the annotated training images, processing image data with the component detectors to identify at least one of the components, and executing the rules to detect the object based on the identified components.
摘要翻译：一种检测被认为是威胁的图像数据中的对象的方法包括：注释至少一个训练图像的部分，以指示每个部分是否是对象的组件，使用多个第一顺序对描述对象的模式语法进行编码基于逻辑的谓词规则，训练不同的分量检测器，以基于所标注的训练图像来识别相应的一个分量，使用分量检测器处理图像数据以识别组件中的至少一个，以及执行规则以检测基于对象对所识别的组件。

6. 发明授权

US09183294B2 Meta-data approach to querying multiple biomedical ontologies 有权
标题翻译：元数据方法查询多个生物医学本体
公开(公告)号：US09183294B2
公开(公告)日：2015-11-10
申请号：US13442124
申请日：2012-04-09
申请人： Ravi Kiran Reddy Palla , Dan G. Tecuci , Vinay Damodar Shet , Mathaeus Dejori
发明人： Ravi Kiran Reddy Palla , Dan G. Tecuci , Vinay Damodar Shet , Mathaeus Dejori
IPC分类号： G06F17/30
CPC分类号： G06F17/30734
摘要： A method for retrieving information spread across a plurality of different ontologies, including: defining a meta-ontology, wherein the meta-ontology includes high-level properties and their mappings to specific properties defined in a plurality of different ontologies; receiving a question, wherein the question is associated with a high-level property; and providing an answer to the question, wherein the answer is determined by using the meta-ontology.
摘要翻译：一种用于检索分布在多个不同本体上的信息的方法，包括：定义元本体，其中所述元本体包括高级属性及其与在多个不同本体中定义的特定属性的映射; 接收问题，其中该问题与高级别财产相关联; 并提供对该问题的答案，其中通过使用元本体来确定答案。

7. 发明授权

US08639678B2 System for generating a medical knowledge base 有权
标题翻译：生成医学知识库的系统
公开(公告)号：US08639678B2
公开(公告)日：2014-01-28
申请号：US13479363
申请日：2012-05-24
申请人： Swapna Somasundaran , Vinodkumar Prabhakaran , Vinay Damodar Shet , Kateryna Tymoshenko , Mathäus Dejori
发明人： Swapna Somasundaran , Vinodkumar Prabhakaran , Vinay Damodar Shet , Kateryna Tymoshenko , Mathäus Dejori
IPC分类号： G06F17/30
CPC分类号： G06F17/30705 , G06F19/00 , G16H50/20 , G16H50/70 , Y10S707/99933
摘要： A system generates medical knowledge base information by searching at least one repository of medical information to identify sentences including a received medical term. A data processor searches the identified sentences to identify sentences including a medical term different to the received term in response to a predetermined repository of medical terms and excludes sentences without a term different to the received term, to provide remaining multiple term sentences. The data processor groups different terms of individual sentences of the multiple term sentences to provide grouped terms, determines whether a medically valid relationship occurs between different terms of an individual group of terms of the grouped terms by using predetermined sentence structure and syntax rules and outputs data representing grouped terms having a medically valid relationship.
摘要翻译：系统通过搜索至少一个医疗信息库来识别包括接收到的医疗术语的句子来生成医学知识库信息。数据处理器搜索所识别的句子以响应于医学术语的预定存储库来识别包括与接收到的术语不同的医学术语的句子，并且排除不具有与接收到的术语不同的术语的句子，以提供剩余的多个句子句子。数据处理器对多个句子的单个句子的不同术语进行分组以提供分组的术语，通过使用预定的句子结构和语法规则来确定分组术语的单个术语组的不同术语之间是否发生医学上有效的关系，并输出数据代表具有医学上有效关系的分组术语。

8. 发明申请

US20130066903A1 System for Linking Medical Terms for a Medical Knowledge Base 有权
标题翻译：连接医疗知识库医学术语的系统
公开(公告)号：US20130066903A1
公开(公告)日：2013-03-14
申请号：US13479388
申请日：2012-05-24
申请人： Kateryna Tymoshenko , Swapna Somasundaran , Vinay Damodar Shet
发明人： Kateryna Tymoshenko , Swapna Somasundaran , Vinay Damodar Shet
IPC分类号： G06F17/30
CPC分类号： G06F17/30705 , G06F19/00 , G16H50/20 , G16H50/70 , Y10S707/99933
摘要： A system generates medical knowledge base information by using predetermined data source specific message syntax information in identifying first and second information received from first and second data sources respectively. The first and second information indicates at least one type of medical relationship between the received first and second medical terms. The system determines likelihood of existence of the at least one type of medical relationship indicated by a combination of the first and second information, in response to predetermined information indicating a number of occurrences of the at least one type of relationship in data of at least one of the first and second data source. The system outputs first and second medical terms and the at least one type of medical relationship in response to the determined likelihood of existence.
摘要翻译：系统通过使用预定的数据源特定消息语法信息分别识别从第一和第二数据源接收的第一和第二信息来生成医学知识库信息。第一和第二信息指示接收到的第一和第二医疗术语之间的至少一种类型的医疗关系。响应于指示至少一种数据中的至少一种类型的关系的出现次数的预定信息，该系统确定存在由第一和第二信息的组合指示的至少一种类型的医疗关系的可能性的第一和第二数据源。该系统响应确定的存在的可能性而输出第一和第二医疗术语和至少一种类型的医疗关系。

9. 发明授权

US08315965B2 Method for object detection 有权
标题翻译：物体检测方法
公开(公告)号：US08315965B2
公开(公告)日：2012-11-20
申请号：US12107151
申请日：2008-04-22
申请人： Vinay Damodar Shet
发明人： Vinay Damodar Shet
IPC分类号： G06F17/00 , G06N5/02
CPC分类号： G06K9/626 , G06K9/00369
摘要： A method for object detection from a visual image of a scene. The method includes: using a first order predicate logic formalism to specify a set of logical rules to encode contextual knowledge regarding the object to be detected; inserting the specified logical rules into a knowledge base; obtaining the visual image of the scene; applying specific object feature detectors to some or all pixels in the visual image of the scene to obtain responses at those locations; using the obtained responses to generate logical facts indicative of whether specific features or parts of the object are present or absent at that location in the visual image; inserting the generated logical facts into the knowledge base; and combining the logical facts with the set of logical rules to whether the object is present or absent at a particular location in the scene.
摘要翻译：一种用于从场景的视觉图像进行物体检测的方法。该方法包括：使用一阶谓词逻辑形式来指定一组逻辑规则以对关于待检测对象的上下文知识进行编码; 将指定的逻辑规则插入到知识库中; 获取场景的视觉图像; 将特定对象特征检测器应用于场景的视觉图像中的一些或所有像素，以在那些位置处获得响应; 使用所获得的响应来生成指示所述对象的特定特征或部分是否存在于所述视觉图像中的所述位置处的逻辑事实; 将生成的逻辑事实插入到知识库中; 以及将所述逻辑事实与所述逻辑规则集合，以组合所述对象是否存在于场景中的特定位置处。

10. 发明授权

US08548231B2 Predicate logic based image grammars for complex visual pattern recognition 有权
标题翻译：用于复杂视觉模式识别的基于逻辑的图像语法
公开(公告)号：US08548231B2
公开(公告)日：2013-10-01
申请号：US12724954
申请日：2010-03-16
申请人： Vinay Damodar Shet , Maneesh Kumar Singh , Claus Bahlmann , Visvanathan Ramesh , Stephen P. Masticola , Jan Neumann , Toufiq Parag , Michael A. Gall , Roberto Antonio Suarez
发明人： Vinay Damodar Shet , Maneesh Kumar Singh , Claus Bahlmann , Visvanathan Ramesh , Stephen P. Masticola , Jan Neumann , Toufiq Parag , Michael A. Gall , Roberto Antonio Suarez
IPC分类号： G06K9/62 , G06F17/00
CPC分类号： G06K9/00369 , G06K9/626
摘要： First order predicate logics are provided, extended with a bilattice based uncertainty handling formalism, as a means of formally encoding pattern grammars, to parse a set of image features, and detect the presence of different patterns of interest implemented on a processor. Information from different sources and uncertainties from detections, are integrated within the bilattice framework. Automated logical rule weight learning in the computer vision domain applies a rule weight optimization method which casts the instantiated inference tree as a knowledge-based neural network, to converge upon a set of rule weights that give optimal performance within the bilattice framework. Applications are in (a) detecting the presence of humans under partial occlusions and (b) detecting large complex man made structures in satellite imagery (c) detection of spatio-temporal human and vehicular activities in video and (c) parsing of Graphical User Interfaces.
摘要翻译：提供了第一级谓词逻辑，扩展了基于双精度的不确定性处理形式，作为正式编码模式语法的手段，解析一组图像特征，并检测在处理器上实现的不同兴趣模式的存在。来自不同来源的信息和来自检测的不确定性，被集成在双层框架内。计算机视觉领域中的自动逻辑规则权重学习应用规则权重优化方法，该方法将实例化推理树作为基于知识的神经网络进行聚合，以收敛于在边界框架内提供最佳性能的一组规则权重。应用是（a）检测部分闭塞下人的存在，（b）检测卫星图像中的大型复杂人造结构（c）检测视频中的时空人类和车辆活动，（c）解析图形用户界面。

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