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. 发明授权

US08239333B2 Media tag recommendation technologies 有权
标题翻译：媒体标签推荐技术
公开(公告)号：US08239333B2
公开(公告)日：2012-08-07
申请号：US12396885
申请日：2009-03-03
申请人： Linjun Yang , Lei Wu , Xian-Sheng Hua
发明人： Linjun Yang , Lei Wu , Xian-Sheng Hua
IPC分类号： G06F15/18
CPC分类号： G06F17/3089 , G06Q10/10
摘要： Technologies for recommending relevant tags for the tagging of media based on one or more initial tags provided for the media and based on a large quantity of other tagged media. Sample media as candidates for recommendation are provided by a set of weak rankers based on corresponding relevance measures in semantic and visual domains. The various samples provided by the weak rankers are then ranked based on relative order to provide a list of recommended tags for the media. The weak rankers provide sample tags based on relevance measures including tag co-occurrence, tag content correlation, and image-conditioned tag correlation.
摘要翻译：基于为媒体提供的一个或多个初始标签并基于大量其他标记的媒体来推荐用于标记媒体的相关标签的技术。作为推荐候选人的示例媒体由一组基于语义和视觉领域中的相应相关性度量的弱排名者提供。然后由弱排名者提供的各种样本根据相关顺序排列，以提供媒体推荐标签的列表。弱排名者基于相关性测量提供样本标签，包括标签共现，标签内容相关性和图像条件标签相关性。

2. 发明授权

US08706674B2 Media tag recommendation technologies 有权
标题翻译：媒体标签推荐技术
公开(公告)号：US08706674B2
公开(公告)日：2014-04-22
申请号：US13537802
申请日：2012-06-29
申请人： Linjun Yang , Lei Wu , Xian-Sheng Hua
发明人： Linjun Yang , Lei Wu , Xian-Sheng Hua
IPC分类号： G06F19/00 , G06T1/00
CPC分类号： G06F17/3089 , G06Q10/10
摘要： Technologies for recommending relevant tags for the tagging of media based on one or more initial tags provided for the media and based on a large quantity of other tagged media. Sample media as candidates for recommendation are provided by a set of weak rankers based on corresponding relevance measures in semantic and visual domains. The various samples provided by the weak rankers are then ranked based on relative order to provide a list of recommended tags for the media. The weak rankers provide sample tags based on relevance measures including tag co-occurrence, tag content correlation, and image-conditioned tag correlation.
摘要翻译：基于为媒体提供的一个或多个初始标签并基于大量其他标记的媒体来推荐用于标记媒体的相关标签的技术。作为推荐候选人的示例媒体由一组基于语义和视觉领域中的相应相关性度量的弱排名者提供。然后由弱排名者提供的各种样本根据相关顺序排列，以提供媒体推荐标签的列表。弱排名者基于相关性测量提供样本标签，包括标签共现，标签内容相关性和图像条件标签相关性。

3. 发明申请

US20120265772A1 MEDIA TAG RECOMMENDATION TECHNOLOGIES 有权
标题翻译：媒体标签推荐技术
公开(公告)号：US20120265772A1
公开(公告)日：2012-10-18
申请号：US13537802
申请日：2012-06-29
申请人： Linjun Yang , Lei Wu , Xian-Sheng Hua
发明人： Linjun Yang , Lei Wu , Xian-Sheng Hua
IPC分类号： G06F17/30
CPC分类号： G06F17/3089 , G06Q10/10
摘要： Technologies for recommending relevant tags for the tagging of media based on one or more initial tags provided for the media and based on a large quantity of other tagged media. Sample media as candidates for recommendation are provided by a set of weak rankers based on corresponding relevance measures in semantic and visual domains. The various samples provided by the weak rankers are then ranked based on relative order to provide a list of recommended tags for the media. The weak rankers provide sample tags based on relevance measures including tag co-occurrence, tag content correlation, and image-conditioned tag correlation.
摘要翻译：基于为媒体提供的一个或多个初始标签并基于大量其他标记的媒体来推荐用于标记媒体的相关标签的技术。作为推荐候选人的示例媒体由一组基于语义和视觉领域中的相应相关性度量的弱排名者提供。然后由弱排名者提供的各种样本根据相关顺序排列，以提供媒体推荐标签的列表。弱排名者基于相关性测量提供样本标签，包括标签共现，标签内容相关性和图像条件标签相关性。

4. 发明申请

US20100228691A1 Media Tag Recommendation Technologies 有权
标题翻译：媒体标签推荐技术
公开(公告)号：US20100228691A1
公开(公告)日：2010-09-09
申请号：US12396885
申请日：2009-03-03
申请人： Linjun Yang , Lei Wu , Xian-Sheng Hua
发明人： Linjun Yang , Lei Wu , Xian-Sheng Hua
IPC分类号： G06F15/18 , G06N5/02 , G06F17/30
CPC分类号： G06F17/3089 , G06Q10/10
摘要： Technologies for recommending relevant tags for the tagging of media based on one or more initial tags provided for the media and based on a large quantity of other tagged media. Sample media as candidates for recommendation are provided by a set of weak rankers based on corresponding relevance measures in semantic and visual domains. The various samples provided by the weak rankers are then ranked based on relative order to provide a list of recommended tags for the media. The weak rankers provide sample tags based on relevance measures including tag co-occurrence, tag content correlation, and image-conditioned tag correlation.
摘要翻译：基于为媒体提供的一个或多个初始标签并基于大量其他标记的媒体来推荐用于标记媒体的相关标签的技术。作为推荐候选人的示例媒体由一组基于语义和视觉领域中的相应相关性度量的弱排名者提供。然后由弱排名者提供的各种样本根据相关顺序排列，以提供媒体推荐标签的列表。弱排名者基于相关性测量提供样本标签，包括标签共现，标签内容相关性和图像条件标签相关性。

5. 发明申请

US20100106486A1 IMAGE-BASED SEMANTIC DISTANCE 有权
标题翻译：基于图像的语义距离
公开(公告)号：US20100106486A1
公开(公告)日：2010-04-29
申请号：US12340632
申请日：2008-12-19
申请人： Xian-Sheng Hua , Lei Wu , Shipeng Li
发明人： Xian-Sheng Hua , Lei Wu , Shipeng Li
IPC分类号： G06F17/27 , G06K9/00
CPC分类号： G06K9/723 , G06K9/6226
摘要： Image-based semantic distance technique embodiments are presented that involve establishing a measure of an image-based semantic distance between semantic concepts. Generally, this entails respectively computing a semantic concept representation for each concept based on a collection of images associated with the concept. A degree of difference is then computed between two semantic concept representations to produce the aforementioned semantic distance measure for the pair of corresponding concepts.
摘要翻译：提出了基于图像的语义距离技术实施例，其涉及建立语义概念之间基于图像的语义距离的度量。通常，这需要基于与该概念相关联的图像的集合来分别计算每个概念的语义概念表示。然后在两个语义概念表示之间计算差异程度，以产生用于该对相应概念的上述语义距离度量。

6. 发明授权

US08645123B2 Image-based semantic distance 有权
标题翻译：基于图像的语义距离
公开(公告)号：US08645123B2
公开(公告)日：2014-02-04
申请号：US12340632
申请日：2008-12-19
申请人： Xian-Sheng Hua , Lei Wu , Shipeng Li
发明人： Xian-Sheng Hua , Lei Wu , Shipeng Li
IPC分类号： G06F17/27 , G06F17/20 , G06F17/21
CPC分类号： G06K9/723 , G06K9/6226
摘要： Image-based semantic distance technique embodiments are presented that involve establishing a measure of an image-based semantic distance between semantic concepts. Generally, this entails respectively computing a semantic concept representation for each concept based on a collection of images associated with the concept. A degree of difference is then computed between two semantic concept representations to produce the aforementioned semantic distance measure for the pair of corresponding concepts.
摘要翻译：提出了基于图像的语义距离技术实施例，其涉及建立语义概念之间基于图像的语义距离的度量。通常，这需要基于与该概念相关联的图像的集合来分别计算每个概念的语义概念表示。然后在两个语义概念表示之间计算差异程度，以产生用于该对相应概念的上述语义距离度量。

7. 发明授权

US08569009B2 Methods and compositions for detecting rare cells from a biological sample 有权
公开(公告)号：US08569009B2
公开(公告)日：2013-10-29
申请号：US11841972
申请日：2007-08-20
申请人： Ping Lin , Andrea Ghetti , Wenge Shi , Mengjia Tang , Gioulnar I. Harvie , Huimin Tao , Guoliang Tao , Lei Wu , David Cerny , Jia Xu
发明人： Ping Lin , Andrea Ghetti , Wenge Shi , Mengjia Tang , Gioulnar I. Harvie , Huimin Tao , Guoliang Tao , Lei Wu , David Cerny , Jia Xu
IPC分类号： C12Q1/24
摘要： Provided are methods and compositions for isolating and detecting rare cells from a biological sample containing other types of cells, particularly including debulking that uses a microfabricated filter for filtering samples. The enriched rare cells can be used in a downstream process such as identification, characterization or growth in culture, or in other ways. Also included is a method of determining tumor aggressiveness or the number or proportion of cancer cells in the enriched sample by detecting telomerase activity, nucleic acid or expression after enrichment of rare cells. Also provided is an efficient, rapid method to specifically remove red and white blood cells from a biological sample containing at least one of the cell types, leading to enrichment of rare target cells including circulating tumor (CTC), stromal, mesenchymal, endothelial, fetal, stem, or non-hematopoietic cells et cetera from a blood sample.

8. 发明授权

US08126274B2 Visual language modeling for image classification 有权
标题翻译：图像分类的视觉语言建模
公开(公告)号：US08126274B2
公开(公告)日：2012-02-28
申请号：US11847959
申请日：2007-08-30
申请人： Mingjing Li , Wei-Ying Ma , Zhiwei Li , Lei Wu
发明人： Mingjing Li , Wei-Ying Ma , Zhiwei Li , Lei Wu
IPC分类号： G06K9/62
CPC分类号： G06K9/4685 , G06K9/4642 , G06K9/6278
摘要： Systems and methods for visual language modeling for image classification are described. In one aspect the systems and methods model training images corresponding to multiple image categories as matrices of visual words. Visual language models are generated from the matrices. In view of a given image, for example, provided by a user or from the Web, the systems and methods determine an image category corresponding to the given image. This image categorization is accomplished by maximizing the posterior probability of visual words associated with the given image over the visual language models. The image category, or a result corresponding to the image category, is presented to the user.
摘要翻译：描述了用于图像分类的视觉语言建模的系统和方法。在一个方面，系统和方法将对应于多个图像类别的训练图像建模为视觉词的矩阵。视觉语言模型是从矩阵生成的。考虑到例如由用户或从Web提供的给定图像，系统和方法确定对应于给定图像的图像类别。这种图像分类是通过在视觉语言模型上最大化与给定图像相关联的视觉词的后验概率来实现的。图像类别或与图像类别对应的结果被呈现给用户。

9. 发明授权

US07835425B1 Architectures, circuits, systems and methods for reducing latency in data communications 有权
标题翻译：用于减少数据通信延迟的架构，电路，系统和方法
公开(公告)号：US07835425B1
公开(公告)日：2010-11-16
申请号：US12330218
申请日：2008-12-08
申请人： Pantas Sutardja , Lei Wu , Hongying Sheng
发明人： Pantas Sutardja , Lei Wu , Hongying Sheng
IPC分类号： H03K11/00 , H04L25/60 , H04L25/64
CPC分类号： G06F13/405
摘要： Circuits, architectures, systems and methods for facilitating data communications and/or reducing latency in data communications. The architecture includes a clock recovery loop receiving data from a host device and providing a recovered clock signal, a filter circuit receiving recovered clock signal information and providing a control signal that adjusts the transmitter clock in response to recovered clock signal information and the two clock signals, and a transmitter receiving the control signal and transmitting data to a destination device in accordance with the transmitter clock. The circuitry generally includes a clock alignment block receiving first and second periodic signals and providing a control signal in response thereto, a filter for first periodic signal information, and a logic circuit configured to combine the control signal and the filtered information, thereby providing an adjustment signal for the second periodic signal. The systems generally relate to those that include the present architecture and/or circuit. The method generally includes determining a phase difference between first and second periodic signals, one of the periodic signals being recovered from a data stream; adjusting the other periodic signal in response to the phase difference and filtered information from the recovered periodic signal; and transmitting the data stream in accordance with said adjusted periodic signal. The present invention advantageously eliminates a FIFO memory in the data path, thereby reducing transceiver latency and improving system performance.
摘要翻译：用于促进数据通信和/或减少数据通信中的延迟的电路，架构，系统和方法。该架构包括时钟恢复环路，其接收来自主机设备的数据并提供恢复的时钟信号，滤波器电路接收恢复的时钟信号信息，并提供响应于恢复的时钟信号信息和两个时钟信号调整发射机时钟的控制信号以及接收控制信号并根据发射机时钟向目的地设备发送数据的发射机。电路通常包括时钟对准块，其接收第一和第二周期信号并响应于此提供控制信号，用于第一周期性信号信息的滤波器以及被配置为组合控制信号和滤波信息的逻辑电路，由此提供调整信号用于第二周期信号。系统通常涉及包括本架构和/或电路的系统。该方法通常包括确定第一和第二周期信号之间的相位差，从数据流中恢复一个周期信号; 响应于来自恢复的周期信号的相位差和滤波信息调整另一周期信号; 以及根据所述调整的周期信号发送数据流。本发明有利地消除数据路径中的FIFO存储器，从而减少收发机等待时间并提高系统性能。

10. 发明授权

US07813536B2 Image measuring apparatus and method, and image measuring system for glomerular filtration rate 有权
标题翻译：图像测量装置和方法，以及用于肾小球滤过率的图像测量系统
公开(公告)号：US07813536B2
公开(公告)日：2010-10-12
申请号：US11636475
申请日：2006-12-11
申请人： Hongfeng Ma , Qiang Di , Lei Wu , Honglei Fu , Longwei Mei , Liang Tian , Songmin Quan , Jun Zhang , Yan Kang
发明人： Hongfeng Ma , Qiang Di , Lei Wu , Honglei Fu , Longwei Mei , Liang Tian , Songmin Quan , Jun Zhang , Yan Kang
IPC分类号： G06K9/00
CPC分类号： G06K9/34 , A61B6/50 , G06K2209/05 , G06T7/0012 , G06T7/11 , G06T2207/10081 , G06T2207/30084
摘要： The present invention discloses an abdominal CT image measuring apparatus and method. The abdominal CT image measuring apparatus includes: an interface unit; a part recognizing unit and a characteristic data computing unit. The present invention can determine the specific region of the part under test with a little amount of computation, by registration and subtraction operation on the two-phase scan images. This is easy to be carried out in computers, thus the computing speed of the characteristic data can be guaranteed and the efficiency can be improved. By the recognizing of the kidney regions and the abdominal aorta region in the present invention, the glomerular filtration rates obtained by applying the key concept of the present invention to the image measuring of glomerular filtration rate can meet the clinical application requirements in both precision and speed.
摘要翻译：本发明公开了一种腹部CT图像测量装置及方法。腹部CT图像测量装置包括：接口单元; 部件识别单元和特征数据计算单元。本发明可以通过对两相扫描图像进行配准和减法操作，通过少量的计算来确定待测零件的特定区域。这在计算机中很容易实现，因此可以保证特征数据的计算速度并提高效率。通过本发明的肾脏区域和腹主动脉区域的识别，通过将本发明的关键概念应用于肾小球滤过率的图像测定而获得的肾小球滤过率可以满足临床应用的精度和速度要求。

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