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

US20070242872A1 System and method for three-dimensional estimation based on image data 有权
标题翻译：基于图像数据的三维估计系统和方法
公开(公告)号：US20070242872A1
公开(公告)日：2007-10-18
申请号：US11594554
申请日：2006-11-06
申请人： Leonid Rudin , Jean-Michel Morel , Pascal Monasse , Frederic Cao
发明人： Leonid Rudin , Jean-Michel Morel , Pascal Monasse , Frederic Cao
IPC分类号： G06K9/00 , G06K9/36
CPC分类号： G06K9/00664 , G06T7/593
摘要： A method and apparatus to extract a dense three-dimensional model of an observed scene or object from one or more sequences of images acquired with an imaging device such as a camera or camcorder, or clusters of cameras and/or camcorders. In some embodiments the method includes capturing an image sequence by a camera moving with a translateral motion, for example moving on a straight path with a fixed orientation. The images in each sequence are captured at regularly spaced sites on the straight path. In some embodiments, a device projects structured light on the observed scene. Multiple image sequences can be acquired, on different cameras on straight paths. The captured image sequences are input to a computer. The 3D structure of the scene viewed by each image sequence is computed. Dense reconstruction from EPIs is performed by interpolating the EPIs after having detecting straight segments on the EPIs. 3D models extracted from each image sequence are combined and merged to create a 3D model of the whole scene.
摘要翻译：一种从使用诸如照相机或摄像机的成像装置或相机和/或摄像机的集群获取的一个或多个图像序列中提取观察场景或物体的致密三维模型的方法和装置。在一些实施例中，该方法包括通过利用平移运动移动的相机捕获图像序列，例如以具有固定取向的直线路径移动。每个序列中的图像在直线路径上的规则间隔位置捕获。在一些实施例中，设备在所观察的场景上投射结构化光。可以在直路上的不同摄像机上获取多个图像序列。所捕获的图像序列被输入到计算机。计算每个图像序列查看的场景的3D结构。通过在EPI上检测到直线段之后内插EPI来执行来自EPI的密集重建。从每个图像序列提取的3D模型被组合并合并以创建整个场景的3D模型。

2. 发明授权

US07587082B1 Object recognition based on 2D images and 3D models 有权
标题翻译：基于2D图像和3D模型的对象识别
公开(公告)号：US07587082B1
公开(公告)日：2009-09-08
申请号：US11356816
申请日：2006-02-17
申请人： Leonid I. Rudin , Jose-Luis Lisani , Pascal Monasse , Jean-Michel Morel
发明人： Leonid I. Rudin , Jose-Luis Lisani , Pascal Monasse , Jean-Michel Morel
IPC分类号： G06K9/00
CPC分类号： G06K9/00208
摘要： Methods and systems for performing object directed recognition based on two-dimensional images and three-dimensional models. Transforms are used to map features of an object to an image seen from a vantage point and to map features of images best seen from the vantage points to three-dimensional models. Mapping of features and images to the three-dimensional model and then to image planes of other images in comparison of corresponding features from the mapped images to original images allows for determination of coherence between the mapped images and the original images.
摘要翻译：基于二维图像和三维模型执行对象定向识别的方法和系统。变换用于将对象的特征映射到从有利位置看到的图像，并将从有利位置看到的图像的特征映射到三维模型。将特征和图像映射到三维模型，然后将其他图像的图像平面与从映射图像到原始图像的对应特征进行比较，可以确定映射图像和原始图像之间的相干性。

3. 发明授权

US08014588B2 System and method for three-dimensional estimation based on image data 有权
标题翻译：基于图像数据的三维估计系统和方法
公开(公告)号：US08014588B2
公开(公告)日：2011-09-06
申请号：US11594554
申请日：2006-11-06
申请人： Leonid I. Rudin , Jean-Michel Morel , Pascal Monasse , Frédéric Cao
发明人： Leonid I. Rudin , Jean-Michel Morel , Pascal Monasse , Frédéric Cao
IPC分类号： G06K9/00
CPC分类号： G06K9/00664 , G06T7/593
摘要： A method and apparatus to extract a dense three-dimensional model of an observed scene or object from one or more sequences of images acquired with an imaging device such as a camera or camcorder, or clusters of cameras and/or camcorders. In some embodiments the method includes capturing an image sequence by a camera moving with a translateral motion, for example moving on a straight path with a fixed orientation. The images in each sequence are captured at regularly spaced sites on the straight path. In some embodiments, a device projects structured light on the observed scene. Multiple image sequences can be acquired, on different cameras on straight paths. The captured image sequences are input to a computer. The 3D structure of the scene viewed by each image sequence is computed. Dense reconstruction from EPIs is performed by interpolating the EPIs after having detecting straight segments on the EPIs. 3D models extracted from each image sequence are combined and merged to create a 3D model of the whole scene.
摘要翻译：一种从使用诸如照相机或摄像机的成像装置或相机和/或摄像机的集群获取的一个或多个图像序列中提取观察场景或物体的致密三维模型的方法和装置。在一些实施例中，该方法包括通过利用平移运动移动的相机捕获图像序列，例如以具有固定取向的直线路径移动。每个序列中的图像在直线路径上的规则间隔位置捕获。在一些实施例中，设备在所观察的场景上投射结构化光。可以在直路上的不同摄像机上获取多个图像序列。所捕获的图像序列被输入到计算机。计算每个图像序列查看的场景的3D结构。通过在EPI上检测到直线段之后内插EPI来执行来自EPI的密集重建。从每个图像序列提取的3D模型被组合并合并以创建整个场景的3D模型。

4. 发明申请

US20110097008A1 METHOD FOR PROCESSING A DIGITAL OBJECT AND RELATED SYSTEM 有权
标题翻译：用于处理数字对象和相关系统的方法
公开(公告)号：US20110097008A1
公开(公告)日：2011-04-28
申请号：US12672497
申请日：2008-08-07
申请人： Frederic Cao , Frederic Guichard , Noura Azzabou , Jean-Michel Morel , Antoni Buades , Bartomeu Coll
发明人： Frederic Cao , Frederic Guichard , Noura Azzabou , Jean-Michel Morel , Antoni Buades , Bartomeu Coll
IPC分类号： G06K9/40
CPC分类号： G06T3/4015 , G06T5/002 , G06T5/20 , G06T2207/10024
摘要： The invention relates to the processing of a digital object that comprises: cancelling the noise of an original object (I) of a first type containing noise in order to obtain a noise-free object (J) of the first type; obtaining an object with a quasi-white noise of the first type from a difference (B) between the original object and the noise-free object; applying to the noise-free object (J) a first processing (t1) that comprises a neighbouring processing for obtaining a transformed object (K) of a second type, the first processing being such that it would structure the noise contained in the original object if it was applied to said original object; applying to the noise object a second white processing (t2) for obtaining a quasi-white transformed noise object (C) of the second type; and inserting into the transformed object (K) the transformed noise object.
摘要翻译：本发明涉及数字物体的处理，包括：消除包含噪声的第一类型的原始物体（I）的噪声，以获得第一类型的无噪声物体（J）; 从原始对象和无噪声对象之间的差（B）获得具有第一类型的准白噪声的对象; 向无噪声对象（J）施加包括用于获得第二类型的变换对象（K）的相邻处理的第一处理（t1），所述第一处理使得将构成包含在原始对象中的噪声如果它被应用于所述原始对象; 向所述噪声对象施加用于获得所述第二类型的准白变换噪声对象（C）的第二白色处理（t2）; 并将变换的对象（K）插入到变换的噪声对象中。

5. 发明申请

US20100209005A1 REGISTRATION AND COMPARISON OF THREE DIMENSIONAL OBJECTS 失效
标题翻译：三维对象的注册和比较
公开(公告)号：US20100209005A1
公开(公告)日：2010-08-19
申请号：US12371473
申请日：2009-02-13
申请人： Leonid I. Rudin , Jose L. Lisani , Jean-Michel Morel
发明人： Leonid I. Rudin , Jose L. Lisani , Jean-Michel Morel
IPC分类号： G06K9/68
CPC分类号： G06K9/00214
摘要： Information of different scans of physical objects may require comparison, for example to determine if the scans are of the same object or if an object has changed, or better information for a three dimensional model may be desired. Different scans of physical objects may be compared by determining lines or planes tangent to a surface at a discrete number of points, registering three dimensional information provided by the scans using the tangent lines or planes, and determining a measure of discrepancy between the surfaces. Three dimensional information of different scans of the same object may also be merged after determining lines or planes tangent to a surface at a discrete number of points and performing registration and merging.
摘要翻译：物理对象的不同扫描的信息可能需要比较，例如以确定扫描是相同的对象还是对象已经改变，或者可能需要更好的三维模型的信息。可以通过在离散点数处确定与表面相切的线或平面来对物理对象的不同扫描进行比较，通过使用切线或平面记录由扫描提供的三维信息，以及确定表面之间的差异度量。在确定与离散点数之间的表面相切的线或平面之后，可以将相同对象的不同扫描的三维信息合并并执行注册和合并。

6. 发明授权

US08687920B2 Method and device for the invariant-affine recognition of shapes 有权
标题翻译：不变式仿射识别形状的方法和装置
公开(公告)号：US08687920B2
公开(公告)日：2014-04-01
申请号：US12993499
申请日：2009-05-18
申请人： Jean-Michel Morel , Guoshen Yu
发明人： Jean-Michel Morel , Guoshen Yu
IPC分类号： G06K9/32 , G06K9/46
CPC分类号： G06K9/6206 , G06K2209/29
摘要： A method for the recognition of objects in at least one digital image includes: a) simulating from the digital image a plurality of digital rotations and at least two digital tilts different from 1 in order to develop a simulated image for each rotation-tilt pair; and b) applying an algorithm generating values that are invariant in translation, rotation and zoom onto the simulated images in order to determine so-called SIF (scale invariant features) local characteristics used for recognizing objects. The SIFT method can be used in step b.
摘要翻译：用于识别至少一个数字图像中的对象的方法包括：a）从数字图像模拟多个数字旋转和与1不同的至少两个数字倾斜，以便为每个旋转 - 倾斜对形成模拟图像; 以及b）应用一种算法来生成在平移，旋转和缩放中不变的值到仿真图像上，以便确定用于识别对象的所谓的SIF（尺度不变特征）局部特征。 SIFT方法可以在步骤b中使用。

7. 发明申请

US20100141804A1 Image data processing method by reducing image noise, and camera integrating means for implementing said method 有权
标题翻译：通过降低图像噪声的图像数据处理方法，以及用于实现所述方法的相机积分装置
公开(公告)号：US20100141804A1
公开(公告)日：2010-06-10
申请号：US11579380
申请日：2005-04-13
申请人： Jean-Michel Morel , Bartomeu Coll , Antonio Buades
发明人： Jean-Michel Morel , Bartomeu Coll , Antonio Buades
IPC分类号： G06K9/62 , G06K9/40 , H04N5/217
CPC分类号： G06T5/50 , G06T5/002 , G06T2207/20076
摘要： The invention concerns image data processing, through noise reduction comprising the following steps: associating a learning zone (ZA) with a reference point (Pref) of the image (IM); for each variable point (PC, PC′,) of the learning zone, evaluating a distance (d, d′,) between: values of points in a first window (f1) of the image, centered on the reference point, and values of points in a second window (f7, f′2,), of similar format as the format of the first window and centered on the variable point; repeating said distance calculation for all the points of the learning zone as successive variable points and estimating an average value to assign to the reference point, said average being weighted on the basis of the distances evaluated for each variable point.
摘要翻译：本发明涉及通过降噪的图像数据处理，包括以下步骤：将学习区域（ZA）与图像（IM）的参考点（Pref）相关联; 对于学习区域的每个可变点（PC，PC'），评估以下之间的距离（d，d'）：以参考点为中心的图像的第一窗口（f1）中的点的值和值在第二个窗口（f7，f'2，）中的点与第一个窗口的格式类似，并以可变点为中心; 将所述学习区域的所有点的所述距离计算重复为连续可变点，并且估计分配给参考点的平均值，所述平均值基于针对每个可变点评估的距离进行加权。

8. 发明授权

US4819488A Apparatus for testing the resistance to cleavage of cardboard tubes 失效
标题翻译：用于测试纸板管裂解阻力的装置
公开(公告)号：US4819488A
公开(公告)日：1989-04-11
申请号：US162318
申请日：1988-01-21
申请人： Jean-Michel Morel
发明人： Jean-Michel Morel
IPC分类号： G01N3/32 , G01N33/34 , G01N3/24
CPC分类号： G01N3/32 , G01N33/346
摘要： An apparatus for testing the resistance to cleavage of cardboard tubes which have previously been cut into test pieces which includes a key mandrel which is mounted free to rotate, at least one sleeve for each test piece having a different outer diameter, the inner diameter of at least one sleeve being approximately equal to the outer diameter of a test piece so that for each test piece there corresponds at least one sleeve intended to house the test piece. The key mandrel is adapted to receive each test piece surrounded by a corresponding sleeve. A drive device and belt are provided to rotatably drive an assembly which includes the key mandrel and a test piece housed in a corresponding sleeve.
摘要翻译： PCT No.PCT / FR87 / 00188 Sec。 371日期：1988年1月21日 102（e）日期1988年1月21日PCT提交1987年5月29日PCT公布公开号WO87 / 07379 日期为1987年12月3日。一种用于测试先前已被切割成试件的纸板管的开裂抗性的装置，其包括安装成自由旋转的关键芯轴，每个测试件的至少一个套筒具有不同的外部至少一个套筒的内径近似等于试件的外径，使得对于每个试样，至少一个套筒用于容纳试件。钥匙心轴适于接收由相应套筒包围的每个测试件。驱动装置和带被设置为可旋转地驱动包括钥匙心轴的组件和容纳在对应套筒中的试件。

9. 发明授权

US08605989B2 Registration and comparison of three dimensional objects in facial imaging 失效
标题翻译：面部成像中三维物体的注册和比较
公开(公告)号：US08605989B2
公开(公告)日：2013-12-10
申请号：US12371473
申请日：2009-02-13
申请人： Leonid I. Rudin , Jose L. Lisani , Jean-Michel Morel
发明人： Leonid I. Rudin , Jose L. Lisani , Jean-Michel Morel
IPC分类号： G06K9/00 , G06K9/62 , G06T15/00 , G06T17/30
CPC分类号： G06K9/00214
摘要： Information of different scans of physical objects may require comparison, for example to determine if the scans are of the same object or if an object has changed, or better information for a three dimensional model may be desired. Different scans of physical objects may be compared by determining lines or planes tangent to a surface at a discrete number of points, registering three dimensional information provided by the scans using the tangent lines or planes, and determining a measure of discrepancy between the surfaces. Three dimensional information of different scans of the same object may also be merged after determining lines or planes tangent to a surface at a discrete number of points and performing registration and merging.
摘要翻译：物理对象的不同扫描的信息可能需要比较，例如以确定扫描是相同的对象还是对象已经改变，或者可能需要更好的三维模型信息。可以通过在离散点数处确定与表面相切的线或平面来对物理对象的不同扫描进行比较，通过使用切线或平面记录由扫描提供的三维信息，以及确定表面之间的差异度量。在确定与离散点数之间的表面相切的线或平面之后，可以将相同对象的不同扫描的三维信息合并并执行注册和合并。

10. 发明授权

US08427559B2 Image data processing method by reducing image noise, and camera integrating means for implementing said method 有权
标题翻译：通过降低图像噪声的图像数据处理方法，以及用于实现所述方法的相机积分装置
公开(公告)号：US08427559B2
公开(公告)日：2013-04-23
申请号：US13530409
申请日：2012-06-22
申请人： Jean-Michel Morel , Bartomeu Coll , Antonio Buades
发明人： Jean-Michel Morel , Bartomeu Coll , Antonio Buades
IPC分类号： H04N5/217 , H04N9/64
CPC分类号： G06T5/50 , G06T5/002 , G06T2207/20076
摘要： The invention concerns image data processing, through noise reduction comprising the following steps: associating a learning zone (ZA) with a reference point (Pref) of the image (IM); for each variable point (PC, PC′) of the learning zone, evaluating a distance (d,d′,) between: values of points in a first window (f1) of the image, centered on the reference point, and values of points in a second window (f2, f2,), of similar format as the format of the first window and centered on the variable point; repeating said distance calculation for all the points of the learning zone as successive variable points and estimating an average value to assign to the reference point, said average being weighted on the basis of the distances evaluated for each variable point.
摘要翻译：本发明涉及通过降噪的图像数据处理，包括以下步骤：将学习区域（ZA）与图像（IM）的参考点（Pref）相关联; 对于学习区域的每个可变点（PC，PC'），评估以下之间的距离（d，d'）：以参考点为中心的图像的第一窗口（f1）中的点的值和在第二个窗口（f2，f2）中的点与第一个窗口的格式类似，并以可变点为中心; 将所述学习区域的所有点的所述距离计算重复为连续可变点，并且估计分配给参考点的平均值，所述平均值基于针对每个可变点评估的距离进行加权。

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