公开(公告)号：US20070098239A1

公开(公告)日：2007-05-03

申请号：US11466477

申请日：2006-08-23

申请人： Shaohua Zhou ,  Jie Shao ,  Dorin Comaniciu ,  Bogdan Georgescu

发明人： Shaohua Zhou ,  Jie Shao ,  Dorin Comaniciu ,  Bogdan Georgescu

IPC分类号： G06K9/00

CPC分类号： G06K9/621 ,  G06K9/00335 ,  G06K9/48 ,  G06K9/6209 ,  G06K2009/487 ,  G06K2209/05 ,  G06T7/251 ,  G06T2207/10132 ,  G06T2207/30048

摘要： A method for generating Pairwise Active Appearance Models (PAAMs) that characterize shape, appearance and motion of an object and using the PAAM to track the motion of an object is disclosed. A plurality of video streams is received. Each video stream includes a series of image frames that depict an object in motion. Each video stream includes an index of identified motion phases that are associated with a motion cycle of the object. For each video stream, a shape of the object is represented by a shape vector. An appearance of an object is represented by an appearance vector. The shape and appearance vectors associated at two consecutive motion phases are concatenated. Paired data for the concatenated shape and appearance vectors is computed. Paired data is computed for each two consecutive motion phases in the motion cycle. A shape subspace is constructed based on the computed paired data. An appearance subspace is constructed based on the computed paired data. A joint subspace is constructed using a combination of the shape subspace and appearance subspace. A PAAM is generated using the joint subspace and the PAAM is stored in a database.

摘要翻译： 公开了一种用于产生表征对象的形状，外观和运动以及使用PAAM跟踪对象的运动的成对活动外观模型（PAAM）的方法。 接收多个视频流。 每个视频流包括描绘运动对象的一系列图像帧。 每个视频流包括与对象的运动周期相关联的所识别的运动相位的索引。 对于每个视频流，对象的形状由形状向量表示。 物体的外观由外观矢量表示。 在两个连续的运动阶段相关联的形状和外观矢量被连接起来。 计算连续形状和外观向量的配对数据。 计算运动周期中每两个连续运动相位的配对数据。 基于计算的配对数据构建形状子空间。 基于计算出的配对数据构建出现子空间。 使用形状子空间和外观子空间的组合构建关节子空间。 使用联合子空间生成PAAM，并将PAAM存储在数据库中。

公开(公告)号：US20160282111A1

公开(公告)日：2016-09-29

申请号：US14984670

申请日：2015-12-30

申请人： KLA-Tencor Corporation

发明人： Laurent Hermans ,  Eric Delfosse ,  Frans Nijs ,  Francis Thissen ,  Johan De Greeve

IPC分类号： G01B21/20 ,  G01B21/04

CPC分类号： G01B21/20 ,  G01B21/30 ,  G06K9/00214 ,  G06K9/6211 ,  G06K2009/487

摘要： A method for shape classification of an object is provided. Shape categories are provided which specify a plane and points therein relative to the object, and also specify at least one limit coordinate for each such point, the limit coordinate defining a boundary in a direction normal to the plane for the shape of the object considered in order for the object to be classified into a respective shape category. The shape categories can be provided by a user, making the method very flexible. The shape categories can in particular be derived from a set of samples of objects representing a shape category to be defined. For classification, the position of a surface of the object is measured at each of the points defined in the shape category, and the result is compared with the corresponding limit coordinate.

摘要翻译： 提供了一种对象的形状分类方法。 提供了形状类别，其指定相对于对象的平面和点，并且还为每个这样的点指定至少一个极限坐标，限定坐标在垂直于平面的方向上定义所考虑的对象的形状 对象被分类为相应的形状类别的顺序。 形状类别可以由用户提供，使得该方法非常灵活。 形状类别可以特别地从表示待定义的形状类别的对象的一组样本导出。 对于分类，在形状类别中定义的每个点处测量对象的表面的位置，并将结果与​​相应的极限坐标进行比较。

公开(公告)号：US20160267352A1

公开(公告)日：2016-09-15

申请号：US15159541

申请日：2016-05-19

申请人： Materialise N.V.

发明人： Joris VAN DEUN

IPC分类号： G06K9/62 ,  G06K9/48

CPC分类号： G06K9/6209 ,  G06K9/48 ,  G06K9/6206 ,  G06K9/6255 ,  G06K2009/487

摘要： Systems and methods are provided for constructing a statistical shape model from a set of training shapes. In one embodiment, two shapes in the training set can be parameterized to a common base domain. Correspondence between the shapes can be evaluated using shape-specific data, such as, for the case of anatomical shapes, anatomical curves and/or anatomical landmarks. In evaluating correspondence, the shape-specific data about the second shape can be mapped to the shape-specific data about the first shape, and the mapping can be optimized based at least in part on a deformation energy of the mapping.

摘要翻译： 提供了从一组训练形状构建统计形状模型的系统和方法。 在一个实施例中，训练集中的两个形状可被参数化为公共基本域。 形状之间的对应可以使用特定形状的数据来评估，例如对于解剖形状，解剖曲线和/或解剖标志的情况。 在评估对应时，关于第二形状的形状特定数据可以映射到关于第一形状的形状特定数据，并且可以至少部分地基于映射的变形能来优化映射。

公开(公告)号：US09393697B1

公开(公告)日：2016-07-19

申请号：US14677486

申请日：2015-04-02

申请人： Disney Enterprises, Inc.

发明人： Paul Beardsley ,  Aparna Taneja

IPC分类号： B25J13/08 ,  B25J11/00 ,  B25J19/02 ,  H04N5/232 ,  G06T1/00

CPC分类号： B25J11/0005 ,  B25J19/021 ,  G06K9/00771 ,  G06K9/4652 ,  G06K2009/487 ,  G06T1/0014 ,  H04N5/23222 ,  H04N5/23229 ,  Y10S901/01 ,  Y10S901/47

摘要： The present disclosure discloses a system and method for creating a customized guest experience at an amusement park. In one example, the method includes capturing by a foot sensor a first foot shape corresponding to at least one foot in a pair of feet of a guest and capturing by a camera a first foot appearance corresponding to at least one foot in the first pair of feet receiving guest data from the guest. The method also includes generating a first foot model using the first foot shape and the first foot appearance and tagging the first foot model with the guest data. The foot model can be used to identify a particular guest and the guest data can be used to output a customized guest experience to the guest.

摘要翻译： 本公开公开了一种用于在游乐园创建定制的客座体验的系统和方法。 在一个示例中，该方法包括由脚传感器捕获与客人的一对脚中的至少一个脚相对应的第一脚形，并且由相机捕获与第一对中的至少一个脚相对应的第一脚部外观 接收来宾的客人数据。 该方法还包括使用第一脚形和第一脚外观生成第一脚模型，并用访客数据标记第一脚模型。 脚型可用于识别特定的访客，访客数据可用于向访客输出定制的访客体验。

公开(公告)号：US20160203387A1

公开(公告)日：2016-07-14

申请号：US14592133

申请日：2015-01-08

申请人： GM GLOBAL TECHNOLOGY OPERATIONS LLC ,  SungKyunKwan University Foundation for Corporate Collaboration

发明人： Sukhan Lee ,  Hung Huu Nguyen ,  Jaewoong Kim ,  Jianying Shi

IPC分类号： G06K9/52 ,  G06T3/00 ,  G06T7/20 ,  H04N13/02 ,  G06T7/00

CPC分类号： G06K9/52 ,  G01B11/00 ,  G01B11/22 ,  G01B11/28 ,  G01C11/00 ,  G01C11/025 ,  G06K9/00201 ,  G06K9/00214 ,  G06K9/46 ,  G06K9/6209 ,  G06K2009/4666 ,  G06K2009/487 ,  G06T3/00 ,  G06T7/00 ,  G06T7/0004 ,  G06T7/13 ,  G06T7/187 ,  G06T7/20 ,  G06T7/50 ,  G06T7/521 ,  G06T7/593 ,  G06T7/60 ,  G06T7/62 ,  G06T7/73 ,  G06T15/00 ,  G06T17/00 ,  G06T17/05 ,  G06T17/20 ,  G06T17/205 ,  G06T2200/04 ,  G06T2200/08 ,  G06T2207/10028 ,  G06T2207/20168 ,  H04N2013/0081 ,  H04N2013/0092

摘要： A vision system is configured to dynamically inspect an object in a field of view. This includes capturing, using a camera, three-dimensional (3D) point cloud data of the field of view and transforming each of the points of the 3D point cloud data into a plurality of tangential surface vectors. Surface normal vectors are determined for each of the points of the 3D point cloud data based upon the plurality of tangential surface vectors. Distribution peaks in the surface normal vectors are detected employing a unit sphere mesh. Parallel planes are separated using the distance distribution peaks. A radially bounded nearest neighbor strategy combined with a process of nearest neighbor searching based upon cell division is executed to segment a planar patch. A planar surface is identified based upon the segmented planar patch.

摘要翻译： 视觉系统被配置为在视野中动态地检查对象。 这包括使用相机捕获视场的三维（3D）点云数据，并将3D点云数据的每个点变换为多个切向表面向量。 基于多个切向表面向量，针对3D点云数据的每个点确定表面法向矢量。 使用单位球网检测表面法向量中的分布峰。 使用距离分布峰分离平行平面。 执行与基于细胞分裂的最近邻搜索的过程组合的径向有界最近邻策略来分割平面补片。 基于分段的平面贴片识别平面。

公开(公告)号：US09305239B2

公开(公告)日：2016-04-05

申请号：US14276433

申请日：2014-05-13

申请人： Samsung Electronics Co., Ltd.

发明人： Shai Litvak

IPC分类号： G06K9/34 ,  G06K9/62 ,  G06T7/00 ,  G06K9/48

CPC分类号： G06K9/6202 ,  G06K9/3266 ,  G06K9/48 ,  G06K9/482 ,  G06K9/6215 ,  G06K2009/487 ,  G06T7/11

摘要： A method for recognizing small-font sized text including receiving digital media of a natural scene, the digital media having at least one frame that includes the small-font sized text; generating input maps having values that reflect local properties of corresponding regions in the at least one frame; and detecting regions of the at least one frame that contain the small-font sized text by integrating information from the input maps. The integrated information may include information located between border lines having active pixels therebetween and gaps having a high ratio of non-ink pixels located below a bottom border line and above a top border line in relation to a dominant direction of the text. The active pixels may be pixels having dense changes in character stroke directions.

摘要翻译： 一种用于识别小字体大小的文本的方法，包括接收自然场景的数字媒体，所述数字媒体具有包括小字体大小的文本的至少一个帧; 生成具有反映所述至少一个帧中的相应区域的局部属性的值的输入图; 以及通过从输入图中整合信息来检测包含小字体大小的文本的至少一个帧的区域。 集成信息可以包括位于其间具有有源像素的边界线之间的信息，以及相对于文本的主导方向位于底部边界线下方和顶部边界线之上的非墨像素的高比率的间隙。 有源像素可以是字符笔划方向上具有密集变化的像素。

公开(公告)号：US20140212028A1

公开(公告)日：2014-07-31

申请号：US13959406

申请日：2013-08-05

申请人： Eagle View Technologies, Inc.

发明人： Christopher Albert Ciarcia

IPC分类号： G06T17/00

CPC分类号： G06T17/05 ,  G06K9/00214 ,  G06K9/48 ,  G06K9/6211 ,  G06K2009/487 ,  G06T7/55 ,  G06T2200/08 ,  G06T2207/10032 ,  G06T2207/20076 ,  G06T2207/30184

摘要： A statistical point pattern matching technique is used to match corresponding points selected from two or more views of a roof of a building. The technique entails statistically selecting points from each of orthogonal and oblique aerial views of a roof, generating radial point patterns for each aerial view, calculating the origin of each point pattern, representing the shape of the point pattern as a radial function, and Fourier-transforming the radial function to produce a feature space plot. A feature profile correlation function can then be computed to relate the point match sets. From the correlation results, a vote occupancy table can be generated to help evaluate the variance of the point match sets, indicating, with high probability, which sets of points are most likely to match one another.

摘要翻译： 使用统计点模式匹配技术来匹配从建筑物屋顶的两个或更多个视图中选出的对应点。 该技术需要统计地从屋顶的正交和倾斜天线视图中选择点，为每个鸟瞰图生成径向点图案，计算每个点图案的原点，将点图案的形状表示为径向函数， 转换径向函数以产生特征空间图。 然后可以计算特征轮廓相关函数以使点匹配集合相关联。 从相关结果可以生成投票占有表，以帮助评估点匹配集的方差，高概率地指示哪一组点最可能相互匹配。

公开(公告)号：US06400828B2

公开(公告)日：2002-06-04

申请号：US09781229

申请日：2001-02-13

申请人： Michele Covell ,  Malcolm Slaney

发明人： Michele Covell ,  Malcolm Slaney

IPC分类号： G06K900

CPC分类号： G06K9/00281 ,  G06K9/4609 ,  G06K9/48 ,  G06K9/6211 ,  G06K2009/487

摘要： The identification of hidden data, such as feature-based control points in an image, from a set of observable data, such as the image, is achieved through a two-stage approach. The first stage involves a learning process, in which a number of sample data sets, e.g. images, are analyzed to identify the correspondence between observable data, such as visual aspects of the image, and the desired hidden data, such as the control points. Two models are created. A feature appearance-only model is created from aligned examples of the feature in the observed data. In addition, each labeled data set is processed to generate a coupled model of the aligned observed data and the associated hidden data. In the second stage of the process, the modeled feature is located in an unmarked, unaligned data set, using the feature appearance-only model. This location is used as an alignment point and the coupled model is then applied to the aligned data, giving an estimate of the hidden data values for that data set.

摘要翻译： 通过两阶段方法，可以从一组可观察数据（如图像）中识别隐藏数据，如图像中基于特征的控制点。 第一阶段涉及学习过程，其中多个样本数据集，例如， 分析图像以识别诸如图像的视觉方面的可观察数据与期望的隐藏数据（例如控制点）之间的对应关系。 创建了两个模型。 仅从观察数据中的特征的对齐示例创建仅出现特征的模型。 此外，处理每个标记的数据集以生成对准的观察数据和相关联的隐藏数据的耦合模型。 在该过程的第二阶段，建模特征位于未标记的未对齐数据集中，使用仅特征外观模型。 该位置用作对齐点，然后将耦合模型应用于对齐的数据，给出该数据集的隐藏数据值的估计。