公开(公告)号：US20060245654A1

公开(公告)日：2006-11-02

申请号：US11119451

申请日：2005-04-29

申请人： Paul Viola ,  Michael Shilman ,  Mukund Narasimhan ,  Percy Liang

发明人： Paul Viola ,  Michael Shilman ,  Mukund Narasimhan ,  Percy Liang

IPC分类号： G06K9/72 ,  G06F7/00

CPC分类号： G06K9/00463 ,  G06K9/726 ,  G06K2209/01

摘要： Grammatical parsing is utilized to parse structured layouts that are modeled as grammars. This type of parsing provides an optimal parse tree for the structured layout based on a grammatical cost function associated with a global search. Machine learning techniques facilitate in discriminatively selecting features and setting parameters in the grammatical parsing process. In one instance, labeled examples are parsed and a chart is generated. The chart is then converted into a subsequent set of labeled learning examples. Classifiers are then trained utilizing conventional machine learning and the subsequent example set. The classifiers are then employed to facilitate scoring of succedent sub-parses. A global reference grammar can also be established to facilitate in completing varying tasks without requiring additional grammar learning, substantially increasing the efficiency of the structured layout analysis techniques.

摘要翻译： 语法解析用于分析模拟为语法的结构化布局。 这种类型的解析为基于与全局搜索相关联的语法成本函数的结构化布局提供了最佳解析树。 机器学习技术有助于在语法解析过程中区分性地选择特征和设置参数。 在一个实例中，已分析标记的示例并生成图表。 然后将该图转换成随后的一组标记的学习示例。 然后使用常规机器学习和随后的示例集训练分类器。 然后使用分类器来方便后续子解析的得分。 还可以建立全局参考语法，以便于完成各种任务，而不需要额外的语法学习，从而大大提高结构化布局分析技术的效率。

公开(公告)号：US20070282538A1

公开(公告)日：2007-12-06

申请号：US11421585

申请日：2006-06-01

申请人： Mukund Narasimhan ,  Paul Viola ,  Michael Shilman

发明人： Mukund Narasimhan ,  Paul Viola ,  Michael Shilman

IPC分类号： G06F19/00

CPC分类号： G06F19/22

摘要： Dynamic inference is leveraged to provide online sequence data labeling. This provides real-time alternatives to current methods of inference for sequence data. Instances estimate an amount of uncertainty in a prediction of labels of sequence data and then dynamically predict a label when an uncertainty in the prediction is deemed acceptable. The techniques utilized to determine when the label can be generated are tunable and can be personalized for a given user and/or a system. Employed decoding techniques can be dynamically adjusted to tradeoff system resources for accuracy. This allows for fine tuning of a system based on available system resources. Instances also allow for online inference because the inference does not require knowledge of a complete set of sequence data.

摘要翻译： 利用动态推理来提供在线序列数据标签。 这提供了对序列数据的推理的当前方法的实时替代。 实例估计序列数据标签的预测中的不确定性量，然后当预测中的不确定性被认为是可接受的时候动态地预测标签。 用于确定何时可以生成标签的技术是可调谐的，并且可以针对给定的用户和/或系统进行个性化。 采用解码技术可以动态调整，以便对系统资源进行权衡以获得准确性。 这允许基于可用的系统资源对系统进行微调。 实例还允许在线推理，因为推理不需要知道一套完整的序列数据。

公开(公告)号：US20060245641A1

公开(公告)日：2006-11-02

申请号：US11119467

申请日：2005-04-29

申请人： Paul Viola ,  Mukund Narasimhan ,  Michael Shilman

发明人： Paul Viola ,  Mukund Narasimhan ,  Michael Shilman

IPC分类号： G06K9/62 ,  G06F17/27 ,  G06K9/46

CPC分类号： G06F17/2715 ,  G06F17/278 ,  G06K9/00469

摘要： A discriminative grammar framework utilizing a machine learning algorithm is employed to facilitate in learning scoring functions for parsing of unstructured information. The framework includes a discriminative context free grammar that is trained based on features of an example input. The flexibility of the framework allows information features and/or features output by arbitrary processes to be utilized as the example input as well. Myopic inside scoring is circumvented in the parsing process because contextual information is utilized to facilitate scoring function training.

摘要翻译： 采用利用机器学习算法的歧视性语法框架来促进用于解析非结构化信息的学习评分功能。 该框架包括基于示例输入的特征进行训练的歧视上下文无关语法。 框架的灵活性允许通过任意进程输出的信息特征和/或特征作为示例输入。 在分析过程中绕过了近视的内分，因为利用上下文信息来促进评分功能训练。

公开(公告)号：US20110314011A1

公开(公告)日：2011-12-22

申请号：US12818377

申请日：2010-06-18

申请人： Greg Buehrer ,  Paul Viola ,  Andrew McGovern ,  Sanaz Ahari ,  Mukund Narasimhan

发明人： Greg Buehrer ,  Paul Viola ,  Andrew McGovern ,  Sanaz Ahari ,  Mukund Narasimhan

IPC分类号： G06F17/30

CPC分类号： G06F16/951

摘要： Computer-readable media, computer systems, and computing devices facilitate generating binary classifier and entity extractor training data. Seed URLs are selected and URL patterns within the seed URLs are identified. Matching URLs in a data structure are identified and corresponding queries and their associated weights are added to a potential training data set from which training data is selected.

摘要翻译： 计算机可读介质，计算机系统和计算设备便于生成二进制分类器和实体提取器训练数据。 选择种子网址，并识别种子网址中的URL模式。 识别数据结构中的匹配URL，并将对应的查询及其相关权重添加到从其中选择训练数据的潜在训练数据集。

公开(公告)号：US20060280370A1

公开(公告)日：2006-12-14

申请号：US11151708

申请日：2005-06-13

申请人： Paul Viola ,  Michael Shilman

发明人： Paul Viola ,  Michael Shilman

IPC分类号： G06K9/72

CPC分类号： G06K9/726 ,  G06F17/271 ,  G06K2209/01

摘要： Image recognition is utilized to facilitate in scoring parse trees for two-dimensional recognition tasks. Trees and subtrees are rendered as images and then utilized to determine parsing scores. Other instances of the subject invention can incorporate additional features such as stroke curvature and/or nearby white space as rendered images as well. Geometric constraints can also be employed to increase performance of a parsing process, substantially improving parsing speed, some even resolvable in polynomial time. Additional performance enhancements can be achieved in yet other instances of the subject invention by employing constellations of integral images and/or integral images of document features.

公开(公告)号：US20060045337A1

公开(公告)日：2006-03-02

申请号：US10927452

申请日：2004-08-26

申请人： Michael Shilman ,  Paul Viola ,  Kumar Chellapilla

发明人： Michael Shilman ,  Paul Viola ,  Kumar Chellapilla

IPC分类号： G06K9/00

CPC分类号： G06K9/726 ,  G06K9/00402 ,  G06K9/344 ,  G06K9/4614 ,  G06K2209/01

摘要： The present invention leverages spatial relationships to provide a systematic means to recognize text and/or graphics. This allows augmentation of a sketched shape with its symbolic meaning, enabling numerous features including smart editing, beautification, and interactive simulation of visual languages. The spatial recognition method obtains a search-based optimization over a large space of possible groupings from simultaneously grouped and recognized sketched shapes. The optimization utilizes a classifier that assigns a class label to a collection of strokes. The overall grouping optimization assumes the properties of the classifier so that if the classifier is scale and rotation invariant the optimization will be as well. Instances of the present invention employ a variant of AdaBoost to facilitate in recognizing/classifying symbols. Instances of the present invention employ dynamic programming and/or A-star search to perform optimization. The present invention applies to both hand-sketched shapes and printed handwritten text, and even heterogeneous mixtures of the two.

摘要翻译： 本发明利用空间关系来提供识别文本和/或图形的系统手段。 这允许以其符号意义来增加草图形状，实现许多功能，包括智能编辑，美化和视觉语言的交互式模拟。 空间识别方法从同时分组和识别的草图形状的可能分组的大空间中获得基于搜索的优化。 优化利用了将类标签分配给笔画集合的分类器。 整体分组优化假设分类器的属性，以便如果分类器是缩放和旋转不变量，则优化将同样如此。 本发明的实施例采用AdaBoost的变体来促进识别/分类符号。 本发明的实施例采用动态规划和/或A星搜索来执行优化。 本发明适用于手绘形状和印刷手写文本，甚至适用于两者的异构混合物。

公开(公告)号：US20070003147A1

公开(公告)日：2007-01-04

申请号：US11173280

申请日：2005-07-01

申请人： Paul Viola ,  Michael Shilman

发明人： Paul Viola ,  Michael Shilman

IPC分类号： G06K9/72 ,  G06K9/34 ,  G06F17/20

CPC分类号： G06K9/726 ,  G06F17/271 ,  G06K2209/01

摘要： A two-dimensional representation of a document is leveraged to extract a hierarchical structure that facilitates recognition of the document. The visual structure is grammatically parsed utilizing two-dimensional adaptations of statistical parsing algorithms. This allows recognition of layout structures (e.g., columns, authors, titles, footnotes, etc.) and the like such that structural components of the document can be accurately interpreted. Additional techniques can also be employed to facilitate document layout recognition. For example, grammatical parsing techniques that utilize machine learning, parse scoring based on image representations, boosting techniques, and/or “fast features” and the like can be employed to facilitate in document recognition.

摘要翻译： 利用文档的二维表示来提取便于识别文档的层次结构。 使用统计解析算法的二维适应来语法解析视觉结构。 这允许识别布局结构（例如，列，作者，标题，脚注等）等，使得可以准确地解释文档的结构组件。 还可以采用附加技术来促进文档布局识别。 例如，可以采用利用机器学习，基于图像表示的分析评分，增强技术和/或“快速特征”等的语法解析技术，以促进文档识别。

公开(公告)号：US20120141020A1

公开(公告)日：2012-06-07

申请号：US13371719

申请日：2012-02-13

申请人： Gang Hua ,  Paul Viola

发明人： Gang Hua ,  Paul Viola

IPC分类号： G06K9/62

CPC分类号： G06F17/3025 ,  G06F17/30262 ,  G06K9/00664 ,  G06K9/4642 ,  G06K9/4652 ,  G06K9/6256 ,  G06K9/6285

摘要： Images are classified as photos (e.g., natural photographs) or graphics (e.g., cartoons, synthetically generated images), such that when searched (online) with a filter, an image database returns images corresponding to the filter criteria (e.g., either photos or graphics will be returned). A set of image statistics pertaining to various visual cues (e.g., color, texture, shape) are identified in classifying the images. These image statistics, combined with pre-tagged image metadata defining an image as either a graphic or a photo, may be used to train a boosting decision tree. The trained boosting decision tree may be used to classify additional images as graphics or photos based on image statistics determined for the additional images.

摘要翻译： 图像被分类为照片（例如，自然照片）或图形（例如，漫画，综合生成的图像）​​，使得当用过滤器搜索（在线）时，图像数据库返回与过滤标准相对应的图像（例如，照片或 图形将被返回）。 在对图像进行分类时，识别关于各种视觉提示（例如，颜色，纹理，形状）的一组图像统计信息。 这些图像统计信息与将图像定义为图形或照片的预先标记的图像元数据可以用于训练增强决策树。 经训练的增强决策树可以用于基于为附加图像确定的图像统计来将附加图像分类为图形或照片。

公开(公告)号：US07822696B2

公开(公告)日：2010-10-26

申请号：US11777471

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06F15/18 ,  G06F17/00

CPC分类号： G06K9/6257 ,  G06K9/00248

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上相同的检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US07619513B2

公开(公告)日：2009-11-17

申请号：US11272095

申请日：2005-11-14

申请人： Maurice L. Hill ,  Michael Mocenter ,  Joeseph S. Reiter ,  Paul Viola ,  Brian Moran

发明人： Maurice L. Hill ,  Michael Mocenter ,  Joeseph S. Reiter ,  Paul Viola ,  Brian Moran

IPC分类号： G08B1/08

CPC分类号： G07C9/00111

摘要： A device for monitoring movement of an object is provided. A first module is configured to secure to the object. A second module, capable of electrically connecting to the first module, includes at least a rechargeable battery and a memory capable of storing a history of movement data. A third module, capable of electrically connecting with the second module, includes a data modem capable of connecting to a remote station, and a battery charger. When the second module is connected to the first module, the memory periodically records available location data representing a position of the device at the time of recording. When the second module is connected to the third module, the memory downloads through the data modem and the battery charger charges the battery.

摘要翻译： 提供了一种用于监视物体的移动的装置。 第一模块被配置为固定到对象。 能够电连接到第一模块的第二模块至少包括可充电电池和能够存储运动数据历史的存储器。 能够与第二模块电连接的第三模块包括能够连接到远程站的数据调制解调器和电池充电器。 当第二模块连接到第一模块时，存储器周期性地记录表示在记录时设备的位置的可用位置数据。 当第二个模块连接到第三个模块时，内存通过数据调制解调器下载，电池充电器为电池充电。