公开(公告)号：US08805104B1

公开(公告)日：2014-08-12

申请号：US13105610

申请日：2011-05-11

申请人： Michele Covell ,  Vivek Kwatra ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja

发明人： Michele Covell ,  Vivek Kwatra ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja

IPC分类号： G06K9/36

CPC分类号： H04N19/97

摘要： An image processing system is provided for encoding images based on example-based compression. The system selects a set of candidate dictionary predictor entries to encode a portion of an image based at least in part on the neighbors of the portion. The spatial continuity between portions of the image is exploited by the image processing system by selecting corresponding dictionary predictor entries that have the same offset vector as the portion of the image and its neighboring portions.

摘要翻译： 提供了一种基于基于实例的压缩来对图像进行编码的图像处理系统。 该系统至少部分地基于该部分的邻居来选择一组候选词典预测变量条目以对图像的一部分进行编码。 图像部分之间的空间连续性由图像处理系统通过选择与图像及其相邻部分具有相同偏移矢量的相应词典预测词条来利用。

公开(公告)号：US08724701B1

公开(公告)日：2014-05-13

申请号：US13088856

申请日：2011-04-18

申请人： Michele Covell ,  Vivek Kwatra ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja

发明人： Michele Covell ,  Vivek Kwatra ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja

IPC分类号： H04N7/12 ,  H04N11/02

CPC分类号： H04N19/23 ,  H04N19/94

摘要： An image processing system is provided for encoding videos based on example-based compression. To select the dictionary predictor entries to encode a video, the image processing system reduces the complexity of the video by decomposing the video into smaller pieces. By breaking the video into the simpler pieces, it is easier to locate dictionary predictor entries that are similar to the pieces of the video. The image processing system may decompose the video into one more space-time tubes. For each space-time tube, the image processing system selects dictionary predictor entries to encode the tube.

摘要翻译： 提供了一种基于基于实例的压缩来对视频进行编码的图像处理系统。 为了选择字典预测器条目来编码视频，图像处理系统通过将视频分解成较小的部分来降低视频的复杂性。 通过将视频分解成更简单的部分，可以更容易地找到类似于视频片段的字典预测变量条目。 图像处理系统可以将视频分解成更多的时空管。 对于每个时空管，图像处理系统选择字典预测器条目来对管进行编码。

公开(公告)号：US08515193B1

公开(公告)日：2013-08-20

申请号：US13083493

申请日：2011-04-08

申请人： Mei Han ,  Michele Covell ,  Saurabh Mathur ,  Shumeet Baluja ,  Vivek Kwatra

发明人： Mei Han ,  Michele Covell ,  Saurabh Mathur ,  Shumeet Baluja ,  Vivek Kwatra

IPC分类号： G06K9/36

CPC分类号： G06K9/6219 ,  G06K9/6807 ,  H04N19/30 ,  H04N19/90

摘要： An exemplar dictionary is built from example image blocks for determining predictor blocks for encoding and decoding images. The exemplar dictionary comprises a hierarchical organization of example image blocks. The hierarchical organization of image blocks is obtained by clustering a set of example image blocks, for example, based on k-means clustering. Performance of clustering is improved by transforming feature vectors representing the image blocks to fewer dimensions. Principal component analysis is used for determining feature vectors with fewer dimensions. The clustering performed at higher levels of the hierarchy uses fewer dimensions of feature vectors compared to lower levels of hierarchy. Performance of clustering is improved by processing only a sample of the image blocks of a cluster. The clustering performed at higher levels of the hierarchy uses lower sampling rates as compared to lower levels of hierarchy.

摘要翻译： 从用于确定用于对图像进行编码和解码的预测器块的示例图像块构建示范字典。 示例性字典包括示例图像块的分级组织。 通过例如基于k均值聚类来聚类一组示例图像块来获得图像块的分级组织。 通过将表示图像块的特征向量变换为较少的维度来提高聚类的性能。 主成分分析用于确定尺寸较小的特征向量。 在层次较高的层次上执行的聚类与较低级别的层次相比，使用较少的特征向量维度。 通过仅处理集群的图像块的样本来提高聚类的性能。 与较低级别的层次相比，在较高层次上执行的聚类使用较低的采样率。

公开(公告)号：US08712930B1

公开(公告)日：2014-04-29

申请号：US13100872

申请日：2011-05-04

申请人： Michele Covell ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja ,  Vivek Kwatra

发明人： Michele Covell ,  Mei Han ,  Saurabh Mathur ,  Shumeet Baluja ,  Vivek Kwatra

IPC分类号： G06F15/18

CPC分类号： G06K9/6202 ,  G06F17/30247 ,  G06K9/00684 ,  G06K9/4604 ,  G06K9/6267 ,  G06K9/6292 ,  G06K9/66 ,  G06T1/0021 ,  H04N19/105 ,  H04N19/17 ,  H04N19/189

摘要： An exemplar dictionary is built from exemplars of digital content for determining predictor blocks for encoding and decoding digital content. The exemplar dictionary organizes the exemplars as clusters of similar exemplars. Each cluster is mapped to a label. Machine learning techniques are used to generate a prediction model for predicting a label for an exemplar. The exemplar dictionary is used to encode digital content. Clusters of exemplars are obtained by applying a prediction model to a target block of digital content for encoding. A predictor block is selected for encoding the target block based on frequency of occurrence of exemplars in the clusters. The target block is encoded using the predictor block.

摘要翻译： 由数字内容的示例构建示范字典，用于确定用于对数字内容进行编码和解码的预测器块。 示范字典将样本组织成类似样本的集群。 每个集群映射到一个标签。 机器学习技术用于生成用于预测样本的标签的预测模型。 示范字典用于对数字内容进行编码。 通过将预测模型应用于用于编码的数字内容的目标块来获得样本簇。 基于群集中的样本的出现频率，选择预测块来对目标块进行编码。 使用预测器块对目标块进行编码。

公开(公告)号：US20150371269A1

公开(公告)日：2015-12-24

申请号：US13598929

申请日：2012-08-30

申请人： Rohan Seth ,  Joshua J. Sacks ,  Michele Covell ,  Michael Chu ,  David Marwood ,  Shumeet Baluja

发明人： Rohan Seth ,  Joshua J. Sacks ,  Michele Covell ,  Michael Chu ,  David Marwood ,  Shumeet Baluja

IPC分类号： G06Q30/02 ,  G06F17/30

CPC分类号： G06Q30/0261 ,  G06F16/9537

摘要： Methods, system, and apparatus, including computer programs are encoded on a computer storage medium, for providing content based on geographic location data. In one aspect, a method includes receiving a request from a website for a map segment corresponding to a particular location. Keywords are extracted from the website as targeting criteria for the particular location. The targeting criteria is associated with the particular location. A subsequent request for advertising content associated with the particular location is received. One or more advertising content items are selected based on the targeting criteria. The selected advertising content items are provided responsive to the subsequent request.

摘要翻译： 包括计算机程序的方法，系统和装置在计算机存储介质上编码，用于基于地理位置数据提供内容。 一方面，一种方法包括从网站接收对应于特定位置的地图段的请求。 将关键字从网站提取为特定位置的定位条件。 定位条件与特定位置相关联。 接收与特定位置相关联的广告内容的后续请求。 基于目标标准来选择一个或多个广告内容项目。 响应于后续请求提供所选择的广告内容项目。

公开(公告)号：US08909631B1

公开(公告)日：2014-12-09

申请号：US13217605

申请日：2011-08-25

申请人： Rohan Seth ,  Joshua J. Sacks ,  Michele Covell ,  Michael Chu ,  Shumeet Baluja ,  David Marwood

发明人： Rohan Seth ,  Joshua J. Sacks ,  Michele Covell ,  Michael Chu ,  Shumeet Baluja ,  David Marwood

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/3087 ,  G06Q30/02

摘要： Methods, systems, and computer program products, including computer programs encoded on a computer readable storage medium, for providing content to a user based on the mode of the user. A method includes: providing a data structure that includes query logs; processing query logs to determine candidate queries that satisfy a criteria, the processing including determining a location of interest, if any, for each query log entry and where the criteria requires an entry to have been submitted from a location that is different from a location of interest associated with the entry; evaluating the candidate queries to determine targeting information for the location of interest; receiving a request for content from a user that is proximate to the location of interest or has expressed an interest in the location of interest; and providing content to the user based at least in part on the targeting information.

摘要翻译： 方法，系统和计算机程序产品，包括在计算机可读存储介质上编码的计算机程序，用于基于用户的模式向用户提供内容。 一种方法包括：提供包括查询日志的数据结构; 处理查询日志以确定满足条件的候选查询，所述处理包括为每个查询日志条目确定感兴趣的位置（如果有的话），并且其中标准要求从不同于位置的位置提交条目 与入场有关的利息; 评估候选查询以确定感兴趣的位置的目标信息; 接收来自接近感兴趣位置的用户的内容请求，或者对感兴趣的位置表示兴趣; 以及至少部分地基于所述目标信息向所述用户提供内容。

公开(公告)号：US08732153B1

公开(公告)日：2014-05-20

申请号：US13616893

申请日：2012-09-14

申请人： Shumeet Baluja ,  Michele Covell

发明人： Shumeet Baluja ,  Michele Covell

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/30646 ,  G06F17/30259 ,  G06F17/30864

摘要： A method includes receiving search queries. Images are generated that represent the volume of each search query relative to all search queries over a period of time. The images are processed to generate wavelets for each image. Related queries having similar increases or decreases in volume are identified by comparing at least some of the wavelets corresponding to a first query to wavelets corresponding to other queries.

摘要翻译： 一种方法包括接收搜索查询。 生成表示在一段时间内相对于所有搜索查询的每个搜索查询的卷的图像。 处理图像以生成每个图像的小波。 通过将对应于第一查询的至少一些小波与对应于其他查询的小波进行比较来识别具有类似体积增加或减小的相关查询。

公开(公告)号：US08515935B1

公开(公告)日：2013-08-20

申请号：US13219076

申请日：2011-08-26

申请人： Shumeet Baluja ,  Michele Covell

发明人： Shumeet Baluja ,  Michele Covell

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/30646 ,  G06F17/30259 ,  G06F17/30864

摘要： A method includes receiving search queries. Images are generated that represent the volume of each search query relative to all search queries over a period of time. The images are processed to generate wavelets for each image. Related queries having similar increases or decreases in volume are identified by comparing at least some of the wavelets corresponding to a first query to wavelets corresponding to other queries.

摘要翻译： 一种方法包括接收搜索查询。 生成表示在一段时间内相对于所有搜索查询的每个搜索查询的卷的图像。 处理图像以生成每个图像的小波。 通过将对应于第一查询的至少一些小波与对应于其他查询的小波进行比较来识别具有类似体积增加或减小的相关查询。

公开(公告)号：US08498951B1

公开(公告)日：2013-07-30

申请号：US13250161

申请日：2011-09-30

申请人： Shumeet Baluja ,  Michele Covell

发明人： Shumeet Baluja ,  Michele Covell

IPC分类号： G06F15/18

CPC分类号： G06N3/08

摘要： A method including training a plurality of learning systems, each learning system implementing a learning function and having an input and producing an output, initializing one or more data structures, and evaluating a target sample is described. Also described are methods that include initializing one or more data structures and evaluating a target sample for a best match.

摘要翻译： 描述了一种训练多个学习系统的方法，每个学习系统实现学习功能并具有输入和产生输出，初始化一个或多个数据结构以及评估目标样本。 还描述了包括初始化一个或多个数据结构并评估目标样本以获得最佳匹配的方法。

公开(公告)号：US08184953B1

公开(公告)日：2012-05-22

申请号：US12391153

申请日：2009-02-23

申请人： Michele Covell ,  Sergey Ioffe ,  Shumeet Baluja ,  David Marwood

发明人： Michele Covell ,  Sergey Ioffe ,  Shumeet Baluja ,  David Marwood

IPC分类号： H04N9/80

CPC分类号： G06F17/30858 ,  G11B27/28

摘要： Videos are received and indexed based on fingerprints generated for the videos, using lookup keys which correspond to sub portions of the reference fingerprints. Specifically, a set of lookup keys is selected such that the number of reference fingerprints such that the clumping associated with the associated each of the lookup keys is minimized. Clumping occurs when a number of reference identifiers associated with a lookup key exceeds a maximum value.

摘要翻译： 基于为视频生成的指纹，使用与参考指纹的子部分相对应的查找键来接收和索引视频。 具体地，选择一组查找键，使得参考指纹的数量使得与相关联的每个查找键相关联的聚集被最小化。 当与查找键相关联的多个参考标识符超过最大值时，会发生分组。