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

US20120207346A1 Detecting and Localizing Multiple Objects in Images Using Probabilistic Inference 有权
标题翻译：使用概率推论检测和定位图像中的多个对象
公开(公告)号：US20120207346A1
公开(公告)日：2012-08-16
申请号：US13024933
申请日：2011-02-10
申请人： Pushmeet Kohli , Victor Lempitsky , Olga Barinova
发明人： Pushmeet Kohli , Victor Lempitsky , Olga Barinova
IPC分类号： G06K9/00
CPC分类号： G06K9/00369 , G06K9/6211
摘要： An object detection system is disclosed herein. The object detection system allows detection of one or more objects of interest using a probabilistic model. The probabilistic model may include voting elements usable to determine which hypotheses for locations of objects are probabilistically valid. The object detection system may apply an optimization algorithm such as a simple greedy algorithm to find hypotheses that optimize or maximize a posterior probability or log-posterior of the probabilistic model or a hypothesis receiving a maximal probabilistic vote from the voting elements in a respective iteration of the algorithm. Locations of detected objects may then be ascertained based on the found hypotheses.
摘要翻译：本文公开了一种对象检测系统。物体检测系统允许使用概率模型来检测一个或多个感兴趣对象。概率模型可以包括可用于确定对象的位置的假设是概率有效的投票元素。对象检测系统可以应用诸如简单贪心算法的优化算法来找到优化或最大化概率模型的后验概率或对数后验的假设，或者从投票元素在相应迭代中接收最大概率投票的假设算法。然后可以基于发现的假设来确定检测到的对象的位置。

2. 发明授权

US08953888B2 Detecting and localizing multiple objects in images using probabilistic inference 有权
标题翻译：使用概率推理检测和定位图像中的多个对象
公开(公告)号：US08953888B2
公开(公告)日：2015-02-10
申请号：US13024933
申请日：2011-02-10
申请人： Pushmeet Kohli , Victor Lempitsky , Olga Barinova
发明人： Pushmeet Kohli , Victor Lempitsky , Olga Barinova
IPC分类号： G06K9/46 , G06K9/00 , G06K9/62
CPC分类号： G06K9/00369 , G06K9/6211
摘要： An object detection system is disclosed herein. The object detection system allows detection of one or more objects of interest using a probabilistic model. The probabilistic model may include voting elements usable to determine which hypotheses for locations of objects are probabilistically valid. The object detection system may apply an optimization algorithm such as a simple greedy algorithm to find hypotheses that optimize or maximize a posterior probability or log-posterior of the probabilistic model or a hypothesis receiving a maximal probabilistic vote from the voting elements in a respective iteration of the algorithm. Locations of detected objects may then be ascertained based on the found hypotheses.
摘要翻译：本文公开了一种对象检测系统。物体检测系统允许使用概率模型来检测一个或多个感兴趣对象。概率模型可以包括可用于确定对象的位置的假设是概率有效的投票元素。对象检测系统可以应用诸如简单贪心算法的优化算法来找到优化或最大化概率模型的后验概率或对数后验的假设，或者从投票元素在相应迭代中接收最大概率投票的假设算法。然后可以基于发现的假设来确定检测到的对象的位置。

3. 发明授权

US08213726B2 Image labeling using multi-scale processing 有权
标题翻译：图像标注使用多尺度处理
公开(公告)号：US08213726B2
公开(公告)日：2012-07-03
申请号：US12488119
申请日：2009-06-19
申请人： Pushmeet Kohli , Carsten Rother , Victor Lempitsky
发明人： Pushmeet Kohli , Carsten Rother , Victor Lempitsky
IPC分类号： G06K9/68
CPC分类号： G06K9/6857 , G06K9/6297 , G06T7/143
摘要： Multi-scale processing may be used to reduce the memory and computational requirements of optimization algorithms for image labeling, for example, for object segmentation, 3D reconstruction, stereo correspondence, optical flow and other applications. For example, in order to label a large image (or 3D volume) a multi-scale process first solves the problem at a low resolution, obtaining a coarse labeling of an original high resolution problem. This labeling is refined by solving another optimization on a subset of the image elements. In examples, an energy function for a coarse level version of an input image is formed directly from an energy function of the input image. In examples, the subset of image elements may be selected using a measure of confidence in the labeling.
摘要翻译：可以使用多尺度处理来减少用于图像标记的优化算法的存储器和计算要求，例如用于对象分割，3D重建，立体声对应，光流等应用。例如，为了标注大图像（或3D体积），多尺度处理首先以低分辨率解决问题，获得原始高分辨率问题的粗略标签。通过在图像元素的子集上求解另一个优化来改进该标记。在示例中，输入图像的粗略级版本的能量函数直接从输入图像的能量函数形成。在示例中，图像元素的子集可以使用标签中置信度的度量来选择。

4. 发明申请

US20100322525A1 Image Labeling Using Multi-Scale Processing 有权
标题翻译：使用多尺度处理的图像标记
公开(公告)号：US20100322525A1
公开(公告)日：2010-12-23
申请号：US12488119
申请日：2009-06-19
申请人： Pushmeet Kohli , Carsten Rother , Victor Lempitsky
发明人： Pushmeet Kohli , Carsten Rother , Victor Lempitsky
IPC分类号： G06K9/68
CPC分类号： G06K9/6857 , G06K9/6297 , G06T7/143
摘要： Multi-scale processing may be used to reduce the memory and computational requirements of optimization algorithms for image labeling, for example, for object segmentation, 3D reconstruction, stereo correspondence, optical flow and other applications. For example, in order to label a large image (or 3D volume) a multi-scale process first solves the problem at a low resolution, obtaining a coarse labeling of an original high resolution problem. This labeling is refined by solving another optimization on a subset of the image elements. In examples, an energy function for a coarse level version of an input image is formed directly from an energy function of the input image. In examples, the subset of image elements may be selected using a measure of confidence in the labeling.
摘要翻译：可以使用多尺度处理来减少用于图像标记的优化算法的存储器和计算要求，例如用于对象分割，3D重建，立体声对应，光流等应用。例如，为了标注大图像（或3D体积），多尺度处理首先以低分辨率解决问题，获得原始高分辨率问题的粗略标签。通过在图像元素的子集上求解另一个优化来改进该标记。在示例中，输入图像的粗略级版本的能量函数直接从输入图像的能量函数形成。在示例中，图像元素的子集可以使用标签中置信度的度量来选择。

5. 发明授权

US09117281B2 Surface segmentation from RGB and depth images 有权
标题翻译： RGB和深度图像的表面分割
公开(公告)号：US09117281B2
公开(公告)日：2015-08-25
申请号：US13287577
申请日：2011-11-02
申请人： Derek Hoiem , Pushmeet Kohli
发明人： Derek Hoiem , Pushmeet Kohli
IPC分类号： G06K9/00 , G06T7/00 , G06T7/60 , H04N13/02 , G06K9/46
CPC分类号： G06T7/0081 , G06K9/00 , G06K9/46 , G06T7/00 , G06T7/11 , G06T7/50 , G06T2207/10024 , G06T2207/10028 , H04N13/20 , H04N13/239
摘要： Surface segmentation from RGB and depth images is described. In one example, a computer receives an image of a scene. The image has pixels which each have an associated color value and an associated depth value representing a distance between from an image sensor to a surface in the scene. The computer uses the depth values to derive a set of three-dimensional planes present within the scene. A cost function is used to determine whether each pixel belongs to one of the planes, and the image elements are labeled accordingly. The cost function has terms dependent on the depth value of a pixel, and the color values of the pixels and at least one neighboring pixel. In various examples, the planes can be extended until they intersect to determine the extent of the scene, and pixels not belonging to a plane can be labeled as objects on the surfaces.
摘要翻译：描述从RGB和深度图像的表面分割。在一个示例中，计算机接收场景的图像。图像具有各自具有相关联的颜色值的像素和表示从图像传感器到场景中的表面之间的距离的相关联的深度值。计算机使用深度值来导出场景中存在的一组三维平面。使用成本函数来确定每个像素是否属于一个平面，并且图像元素被相应地标记。成本函数具有取决于像素的深度值以及像素和至少一个相邻像素的颜色值的项。在各种示例中，平面可以延伸直到它们相交以确定场景的范围，并且不属于平面的像素可以被标记为表面上的对象。

6. 发明授权

US08571263B2 Predicting joint positions 有权
标题翻译：预测联合职位
公开(公告)号：US08571263B2
公开(公告)日：2013-10-29
申请号：US13050858
申请日：2011-03-17
申请人： Jamie Daniel Joseph Shotton , Pushmeet Kohli , Ross Brook Girshick , Andrew Fitzgibbon , Antonio Criminisi
发明人： Jamie Daniel Joseph Shotton , Pushmeet Kohli , Ross Brook Girshick , Andrew Fitzgibbon , Antonio Criminisi
IPC分类号： G06K9/00
CPC分类号： G06F3/017 , G06K9/00362 , G06N5/025
摘要： Predicting joint positions is described, for example, to find joint positions of humans or animals (or parts thereof) in an image to control a computer game or for other applications. In an embodiment image elements of a depth image make joint position votes so that for example, an image element depicting part of a torso may vote for a position of a neck joint, a left knee joint and a right knee joint. A random decision forest may be trained to enable image elements to vote for the positions of one or more joints and the training process may use training images of bodies with specified joint positions. In an example a joint position vote is expressed as a vector representing a distance and a direction of a joint position from an image element making the vote. The random decision forest may be trained using a mixture of objectives.
摘要翻译：例如，描述关节位置的描述是为了在图像中找到人或动物（或其部分）的联合位置，以控制计算机游戏或用于其他应用。在一个实施例中，深度图像的图像元素进行联合位置投票，使得例如描绘躯干的一部分的图像元素可以投射颈部关节，左膝关节和右膝关节的位置。可以对随机决策林进行训练，以使图像元素能够对一个或多个关节的位置进行投票，并且训练过程可以使用具有指定关节位置的身体的训练图像。在一个例子中，联合立场表决被表示为表示从投票的图像元素的联合位置的距离和方向的向量。可以使用目标混合来训练随机决策林。

7. 发明授权

US08401225B2 Moving object segmentation using depth images 有权
标题翻译：使用深度图像移动物体分割
公开(公告)号：US08401225B2
公开(公告)日：2013-03-19
申请号：US13017626
申请日：2011-01-31
申请人： Richard Newcombe , Shahram Izadi , Otmar Hilliges , David Kim , David Molyneaux , Jamie Daniel Joseph Shotton , Pushmeet Kohli , Andrew Fitzgibbon , Stephen Edward Hodges , David Alexander Butler
发明人： Richard Newcombe , Shahram Izadi , Otmar Hilliges , David Kim , David Molyneaux , Jamie Daniel Joseph Shotton , Pushmeet Kohli , Andrew Fitzgibbon , Stephen Edward Hodges , David Alexander Butler
IPC分类号： G06K9/00
CPC分类号： G06T7/215 , G06T7/194 , G06T2207/10028 , G06T2207/30244
摘要： Moving object segmentation using depth images is described. In an example, a moving object is segmented from the background of a depth image of a scene received from a mobile depth camera. A previous depth image of the scene is retrieved, and compared to the current depth image using an iterative closest point algorithm. The iterative closest point algorithm includes a determination of a set of points that correspond between the current depth image and the previous depth image. During the determination of the set of points, one or more outlying points are detected that do not correspond between the two depth images, and the image elements at these outlying points are labeled as belonging to the moving object. In examples, the iterative closest point algorithm is executed as part of an algorithm for tracking the mobile depth camera, and hence the segmentation does not add substantial additional computational complexity.
摘要翻译：描述使用深度图像来移动物体分割。在一个示例中，从从移动深度相机接收的场景的深度图像的背景中分割移动物体。检索场景的先前深度图像，并使用迭代最近点算法与当前深度图像进行比较。迭代最近点算法包括对当前深度图像和先前深度图像之间对应的一组点的确定。在确定点集合期间，检测到一个或多个在两个深度图像之间不对应的离开点，并且将这些离散点处的图像元素标记为属于移动对象。在示例中，迭代最近点算法作为用于跟踪移动深度相机的算法的一部分被执行，因此分割不会增加实质的额外的计算复杂度。

8. 发明申请

US20120237127A1 Grouping Variables for Fast Image Labeling 有权
标题翻译：用于快速图像标记的分组变量
公开(公告)号：US20120237127A1
公开(公告)日：2012-09-20
申请号：US13046967
申请日：2011-03-14
申请人： Pushmeet Kohli , Sebastian Reinhard Bernhard Nowozin
发明人： Pushmeet Kohli , Sebastian Reinhard Bernhard Nowozin
IPC分类号： G06K9/34
CPC分类号： G06K9/00624
摘要： This application describes grouping variables together to minimize cost or time of performing computer vision analysis techniques on images. In one instance, the pixels of an image are represented by a lattice structure of nodes that are connected to each other by edges. The nodes are grouped or merged together based in part on the energy function associated with each edge that connects the nodes together. The energy function of the edge is based in part on the energy functions associated with each node. The energy functions of the node are based on the possible states in which the node may exist. The states of the node are representative of an object, image, or any other feature or classification that may be associated with the pixels in the image.
摘要翻译：该应用程序将变量分组在一起，以最小化对图像执行计算机视觉分析技术的成本或时间。在一个实例中，图像的像素由通过边缘彼此连接的节点的网格结构来表示。部分地基于与将节点连接在一起的每个边缘相关联的能量函数将节点分组或合并在一起。边缘的能量函数部分地基于与每个节点相关联的能量函数。节点的能量函数基于可能存在节点的可能状态。节点的状态表示可能与图像中的像素相关联的对象，图像或任何其他特征或分类。

9. 发明授权

US09626766B2 Depth sensing using an RGB camera 有权
公开(公告)号：US09626766B2
公开(公告)日：2017-04-18
申请号：US14193079
申请日：2014-02-28
申请人： Antonio Criminisi , Duncan Paul Robertson , Peter Kontschieder , Pushmeet Kohli , Henrik Turbell , Adriana Dumitras , Indeera Munasinghe , Jamie Daniel Joseph Shotton
发明人： Antonio Criminisi , Duncan Paul Robertson , Peter Kontschieder , Pushmeet Kohli , Henrik Turbell , Adriana Dumitras , Indeera Munasinghe , Jamie Daniel Joseph Shotton
IPC分类号： G06K9/00 , G06T7/00
CPC分类号： G06T7/50 , G06T2207/10016 , G06T2207/10024 , G06T2207/20072 , G06T2207/20081
摘要： A method of sensing depth using an RGB camera. In an example method, a color image of a scene is received from an RGB camera. The color image is applied to a trained machine learning component which uses features of the image elements to assign all or some of the image elements a depth value which represents the distance between the surface depicted by the image element and the RGB camera. In various examples, the machine learning component comprises one or more entangled geodesic random decision forests.

10. 发明授权

US09053571B2 Generating computer models of 3D objects 有权
标题翻译：生成3D对象的计算机模型
公开(公告)号：US09053571B2
公开(公告)日：2015-06-09
申请号：US13154288
申请日：2011-06-06
申请人： Jamie Daniel Joseph Shotton , Shahram Izadi , Otmar Hilliges , David Kim , David Molyneaux , Pushmeet Kohli , Andrew Fitzgibbon , Stephen Edward Hodges
发明人： Jamie Daniel Joseph Shotton , Shahram Izadi , Otmar Hilliges , David Kim , David Molyneaux , Pushmeet Kohli , Andrew Fitzgibbon , Stephen Edward Hodges
IPC分类号： G06T17/00 , G06T15/00 , G06T17/10 , G06T7/20
CPC分类号： G06T7/251 , G06T17/10 , G06T2200/08 , G06T2207/10028
摘要： Generating computer models of 3D objects is described. In one example, depth images of an object captured by a substantially static depth camera are used to generate the model, which is stored in a memory device in a three-dimensional volume. Portions of the depth image determined to relate to the background are removed to leave a foreground depth image. The position and orientation of the object in the foreground depth image is tracked by comparison to a preceding depth image, and the foreground depth image is integrated into the volume by using the position and orientation to determine where to add data derived from the foreground depth image into the volume. In examples, the object is hand-rotated by a user before the depth camera. Hands that occlude the object are integrated out of the model as they do not move in sync with the object due to re-gripping.
摘要翻译：描述生成3D对象的计算机模型。在一个示例中，使用由基本上静态的深度相机拍摄的对象的深度图像来生成存储在三维体积中的存储器设备中的模型。确定与背景相关的深度图像的部分被去除以留下前景深度图像。通过与前一个深度图像进行比较来跟踪前景深度图像中的对象的位置和方向，并且通过使用位置和方向来将前景深度图像集成到卷中，以确定在哪里添加从前景深度图像导出的数据进入卷。在示例中，该对象在深度相机之前由用户手动旋转。闭合对象的手从模型中集成出来，因为它们不会因为重新抓取而与对象同步移动。

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