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

US20140306874A1 NEAR-PLANE SEGMENTATION USING PULSED LIGHT SOURCE 有权
标题翻译：使用脉冲光源的近平面分割
公开(公告)号：US20140306874A1
公开(公告)日：2014-10-16
申请号：US13862196
申请日：2013-04-12
申请人： Mark Finocchio , Alexandru Balan , Nathan Ackerman , Jeffrey Margolis
发明人： Mark Finocchio , Alexandru Balan , Nathan Ackerman , Jeffrey Margolis
IPC分类号： G06F3/01
CPC分类号： G06F3/017 , G06F3/011
摘要： Methods for recognizing gestures within a near-field environment are described. In some embodiments, a mobile device, such as a head-mounted display device (HMD), may capture a first image of an environment while illuminating the environment using an IR light source with a first range (e.g., due to the exponential decay of light intensity) and capture a second image of the environment without illumination. The mobile device may generate a difference image based on the first image and the second image in order to eliminate background noise due to other sources of IR light within the environment (e.g., due to sunlight or artificial light sources). In some cases, object and gesture recognition techniques may be applied to the difference image in order to detect the performance of hand and/or finger gestures by an end user of the mobile device within a near-field environment of the mobile device.
摘要翻译：描述在近场环境中识别手势的方法。在一些实施例中，诸如头戴式显示设备（HMD）的移动设备可以使用具有第一范围的IR光源（例如，由于指示衰减的指数衰减而捕获环境的第一图像，同时照亮环境光强度），并捕获环境的第二个图像而无需照明。移动设备可以基于第一图像和第二图像生成差异图像，以消除由于环境中的其他IR光源（例如，由于阳光或人造光源）引起的背景噪声。在一些情况下，可以将对象和手势识别技术应用于差异图像，以便在移动设备的近场环境内检测移动设备的最终用户的手和/或手指手势的性能。

2. 发明授权

US08553939B2 Pose tracking pipeline 有权
公开(公告)号：US08553939B2
公开(公告)日：2013-10-08
申请号：US13408775
申请日：2012-02-29
申请人： Robert Matthew Craig , Tommer Leyvand , Craig Peeper , Momin M. Al-Ghosien , Matt Bronder , Oliver Williams , Ryan M. Geiss , Jamie Daniel Joseph Shotton , Johnny Lee , Mark Finocchio
发明人： Robert Matthew Craig , Tommer Leyvand , Craig Peeper , Momin M. Al-Ghosien , Matt Bronder , Oliver Williams , Ryan M. Geiss , Jamie Daniel Joseph Shotton , Johnny Lee , Mark Finocchio
IPC分类号： G06K9/00 , H04N11/02
CPC分类号： G06K9/00335 , A63F13/428 , A63F2300/1006 , A63F2300/1093 , G06F3/011 , G06F3/033 , G06K9/00369 , G06T17/10 , G06T19/00
摘要： A method of tracking a target includes receiving from a source a depth image of a scene including the human subject. The depth image includes a depth for each of a plurality of pixels. The method further includes identifying pixels of the depth image that belong to the human subject and deriving from the identified pixels of the depth image one or more machine readable data structures representing the human subject as a body model including a plurality of shapes.

3. 发明授权

US08295546B2 Pose tracking pipeline 有权
标题翻译：姿势跟踪管道
公开(公告)号：US08295546B2
公开(公告)日：2012-10-23
申请号：US12603437
申请日：2009-10-21
申请人： Robert Matthew Craig , Tommer Leyvand , Craig Peeper , Momin M. Al-Ghosien , Matt Bronder , Oliver Williams , Ryan M. Geiss , Jamie Daniel Joseph Shotton , Johnny Lee , Mark Finocchio
发明人： Robert Matthew Craig , Tommer Leyvand , Craig Peeper , Momin M. Al-Ghosien , Matt Bronder , Oliver Williams , Ryan M. Geiss , Jamie Daniel Joseph Shotton , Johnny Lee , Mark Finocchio
IPC分类号： G06K9/00 , H04N11/02
CPC分类号： G06K9/00335 , A63F13/428 , A63F2300/1006 , A63F2300/1093 , G06F3/011 , G06F3/033 , G06K9/00369 , G06T17/10 , G06T19/00
摘要： A method of tracking a target includes receiving from a source an observed depth image of a scene including the target. Each pixel of the observed depth image is labeled as either a foreground pixel belonging to the target or a background pixel not belonging to the target. Each foreground pixel is labeled with body part information indicating a likelihood that that foreground pixel belongs to one or more body parts of the target. The target is modeled with a skeleton including a plurality of skeletal points, each skeletal point including a three dimensional position derived from body part information of one or more foreground pixels.
摘要翻译：跟踪目标的方法包括从源接收包括目标的场景的观测深度图像。观察到的深度图像的每个像素被标记为属于目标的前景像素或不属于目标的背景像素。每个前景像素用身体部分信息标记，指示该前景像素属于目标的一个或多个身体部位的可能性。目标用包括多个骨骼点的骨架建模，每个骨骼点包括从一个或多个前景像素的身体部位信息导出的三维位置。

4. 发明申请

US20100195867A1 VISUAL TARGET TRACKING USING MODEL FITTING AND EXEMPLAR 审中-公开
标题翻译：使用模型配置和示例的视觉目标跟踪
公开(公告)号：US20100195867A1
公开(公告)日：2010-08-05
申请号：US12367435
申请日：2009-02-06
申请人： Alex Kipman , Mark Finocchio , Ryan M. Geiss , Johnny Chung Lee , Charles Claudius Marais , Zsolt Mathe
发明人： Alex Kipman , Mark Finocchio , Ryan M. Geiss , Johnny Chung Lee , Charles Claudius Marais , Zsolt Mathe
IPC分类号： G06K9/00
CPC分类号： G06K9/00362 , A63F13/10 , A63F13/213 , A63F13/428 , A63F2300/1012 , A63F2300/1093 , A63F2300/6045 , G06F3/01 , G06K9/00369 , G06K9/32 , G06T7/251 , G06T7/285 , G06T2207/10016 , G06T2207/10028 , G06T2207/30196
摘要： A method of tracking a target includes receiving an observed depth image of the target from a source and analyzing the observed depth image with a prior-trained collection of known poses to find an exemplar pose that represents an observed pose of the target. The method further includes rasterizing a model of the target into a synthesized depth image having a rasterized pose and adjusting the rasterized pose of the model into a model-fitting pose based, at least in part, on differences between the observed depth image and the synthesized depth image. Either the exemplar pose or the model-fitting pose is then selected to represent the target.
摘要翻译：跟踪目标的方法包括从源接收观察到的目标深度图像并且利用已知姿势的先前训练集合来分析所观察的深度图像，以找到表示目标观察姿势的示范姿态。该方法还包括将目标的模型光栅化成具有光栅化姿态的合成深度图像，并且至少部分地基于观察到的深度图像和合成的图像之间的差异将模型的光栅化姿态调整为模型拟合姿势深度图像。然后选择示范姿势或模型拟合姿势来表示目标。

5. 发明申请

US20050114293A1 Property management mechanisms for properties in an on-demand property system 有权
标题翻译：物业管理机制的物业在按需物业系统
公开(公告)号：US20050114293A1
公开(公告)日：2005-05-26
申请号：US10939296
申请日：2004-09-11
申请人： Namita Gupta , Mark Finocchio , Michael Hillberg , Jeffrey Bogdan
发明人： Namita Gupta , Mark Finocchio , Michael Hillberg , Jeffrey Bogdan
IPC分类号： G06F7/00 , G06F17/00
CPC分类号： G06F8/24 , Y10S707/99944 , Y10S707/99953
摘要： The techniques and mechanisms described herein are directed at property management mechanisms that operate within an on-demand property system. The property management mechanisms support the caching of property values on an as needed basis, allow a plurality of object instances to utilize the same property via attached properties, overriding default information associated with a property on a per type basis, obtaining values for a property from an external source, and the like. In addition, the on-demand property system provides validation and control for each property instance. The property management mechanisms minimize the storage requirements and provide flexibility without requiring additional code from the developers.
摘要翻译：本文描述的技术和机制针对在按需属性系统内操作的物业管理机制。属性管理机制根据需要支持对属性值的缓存，允许多个对象实例通过附加属性利用相同的属性，覆盖与每个类型基础上的属性关联的默认信息，获取属性值外部源等。此外，按需属性系统为每个属性实例提供验证和控制。物理管理机制可最大限度地减少存储需求并提供灵活性，而无需开发人员附加的代码。

6. 发明授权

US09171264B2 Parallel processing machine learning decision tree training 有权
标题翻译：并行处理机器学习决策树训练
公开(公告)号：US09171264B2
公开(公告)日：2015-10-27
申请号：US12969112
申请日：2010-12-15
申请人： Mark Finocchio , Richard E. Moore , Ryan M. Geiss , Jamie Shotton
发明人： Mark Finocchio , Richard E. Moore , Ryan M. Geiss , Jamie Shotton
IPC分类号： G06F15/80 , G06F15/167 , G06N99/00 , G06N5/04 , G06K9/00 , G06K9/62
CPC分类号： G06N99/005 , G06K9/00986 , G06K9/6281 , G06N5/046
摘要： Embodiments are disclosed herein that relate to generating a decision tree through graphical processing unit (GPU) based machine learning. For example, one embodiment provides a method including, for each level of the decision tree: performing, at each GPU of the parallel processing pipeline, a feature test for a feature in a feature set on every example in an example set. The method further includes accumulating results of the feature tests in local memory blocks. The method further includes writing the accumulated results from each local memory block to global memory to generate a histogram of features for every node in the level, and for each node in the level, assigning a feature having a lowest entropy in accordance with the histograms to the node.
摘要翻译：本文公开的实施例涉及通过基于图形处理单元（GPU）的机器学习来生成决策树。例如，一个实施例提供一种方法，其包括对于决策树的每个级别：在并行处理流水线的每个GPU处执行在示例集合中的每个示例上的特征集中的特征的特征测试。该方法还包括在本地存储器块中积累特征测试的结果。该方法还包括将来自每个本地存储器块的累积结果写入全局存储器以产生该级别中的每个节点的特征的直方图，并且对于该级别中的每个节点，根据直方图将具有最低熵的特征分配到节点。

7. 发明授权

US08565485B2 Pose tracking pipeline 有权
公开(公告)号：US08565485B2
公开(公告)日：2013-10-22
申请号：US13615269
申请日：2012-09-13
申请人： Robert Matthew Craig , Tommer Leyvand , Craig Peeper , Momin M. Al-Ghosien , Matt Bronder , Oliver Williams , Ryan M. Geiss , Jamie Daniel Joseph Shotton , Johnny Lee , Mark Finocchio
发明人： Robert Matthew Craig , Tommer Leyvand , Craig Peeper , Momin M. Al-Ghosien , Matt Bronder , Oliver Williams , Ryan M. Geiss , Jamie Daniel Joseph Shotton , Johnny Lee , Mark Finocchio
IPC分类号： G06K9/00 , H04N11/02
CPC分类号： G06K9/00335 , A63F13/428 , A63F2300/1006 , A63F2300/1093 , G06F3/011 , G06F3/033 , G06K9/00369 , G06T17/10 , G06T19/00
摘要： A method of tracking a target includes receiving from a source a depth image of a scene including the human subject. The depth image includes a depth for each of a plurality of pixels. The method further includes identifying pixels of the depth image that belong to the human subject and deriving from the identified pixels of the depth image one or more machine readable data structures representing the human subject as a body model including a plurality of shapes.

8. 发明申请

US20130141597A1 CONTROLLING POWER CONSUMPTION IN OBJECT TRACKING PIPELINE 有权
标题翻译：控制对象跟踪管道中的功耗
公开(公告)号：US20130141597A1
公开(公告)日：2013-06-06
申请号：US13312846
申请日：2011-12-06
申请人： Kyungsuk David Lee , Mark Finocchio , Richard Moore , Alexandru Balan , Rod G. Fleck
发明人： Kyungsuk David Lee , Mark Finocchio , Richard Moore , Alexandru Balan , Rod G. Fleck
IPC分类号： H04N5/225 , H04N5/228
CPC分类号： G06T7/2046 , G06K9/00362 , G06T7/194 , G06T7/251 , G06T2207/10016 , G06T2207/20224 , G06T2207/30196 , H04N5/23219 , H04N5/23241 , H04N5/23254 , H04N5/2354 , H04N5/345
摘要： Embodiments related to detecting object information from image data collected by an image sensor are disclosed. In one example embodiment, the object information is detected by receiving a frame of image data from the image sensor and detecting a change in a threshold condition related to an object within the frame. The embodiment further comprises adjusting a setting that changes a power consumption of the image sensor in response to detecting the threshold condition.
摘要翻译：公开了从由图像传感器收集的图像数据检测对象信息的实施例。在一个示例实施例中，通过从图像传感器接收图像数据的帧并检测与帧内的对象相关的阈值条件的变化来检测对象信息。该实施例还包括响应于检测阈值条件来调整改变图像传感器的功耗的设置。

9. 发明申请

US20120154373A1 PARALLEL PROCESSING MACHINE LEARNING DECISION TREE TRAINING 有权
标题翻译：平行加工机器学习决策树培训
公开(公告)号：US20120154373A1
公开(公告)日：2012-06-21
申请号：US12969112
申请日：2010-12-15
申请人： Mark Finocchio , Richard E. Moore , Ryan M. Geiss , Jamie Shotton
发明人： Mark Finocchio , Richard E. Moore , Ryan M. Geiss , Jamie Shotton
IPC分类号： G06F15/80 , G06T15/00
CPC分类号： G06N99/005 , G06K9/00986 , G06K9/6281 , G06N5/046
摘要： Embodiments are disclosed herein that relate to generating a decision tree through graphical processing unit (GPU) based machine learning. For example, one embodiment provides a method including, for each level of the decision tree: performing, at each GPU of the parallel processing pipeline, a feature test for a feature in a feature set on every example in an example set. The method further includes accumulating results of the feature tests in local memory blocks. The method further includes writing the accumulated results from each local memory block to global memory to generate a histogram of features for every node in the level, and for each node in the level, assigning a feature having a lowest entropy in accordance with the histograms to the node.
摘要翻译：本文公开的实施例涉及通过基于图形处理单元（GPU）的机器学习来生成决策树。例如，一个实施例提供一种方法，其包括对于决策树的每个级别：在并行处理流水线的每个GPU处执行在示例集合中的每个示例上的特征集中的特征的特征测试。该方法还包括在本地存储器块中积累特征测试的结果。该方法还包括将来自每个本地存储器块的累积结果写入全局存储器以产生该级别中的每个节点的特征的直方图，并且对于该级别中的每个节点，根据直方图将具有最低熵的特征分配到节点。

10. 发明申请

US20060036943A1 Method and system for selectively enforcing presentation themes 审中-公开
标题翻译：有选择地执行演示主题的方法和系统
公开(公告)号：US20060036943A1
公开(公告)日：2006-02-16
申请号：US11258755
申请日：2005-10-25
申请人： Nicholas Kramer , Mark Finocchio , Robert Sinclair
发明人： Nicholas Kramer , Mark Finocchio , Robert Sinclair
IPC分类号： G06F17/24 , G06F3/00
CPC分类号： G06F9/451
摘要： Disclosed is a theme property sheet system that supports presentation themes by coordinating values provided to elements used in a user interface. When an interface user expresses a preference for a particular presentation theme, that theme's property sheet is given precedence over other mechanisms for setting property values. By appropriately setting these values in the interface, the theme property sheet enhances the coherence of the interface. The theme property sheet allows an interface designer to add new presentation elements to a user interface without being concerned about presentation themes. The default for each property as set by the designer is overridden, when necessary, by the theme sheet. In a few particular cases, an application designer needs to override even the user's theme preferences. A mechanism is provided to support these exceptional cases.
摘要翻译：公开了通过协调提供给用户界面中使用的元素的值来支持演示主题的主题属性表系统。当界面用户表达对特定演示主题的偏好时，该主题的属性表优先于设置属性值的其他机制。通过在界面中适当设置这些值，主题属性表可以增强界面的一致性。主题属性表允许界面设计者向用户界面添加新的演示元素，而不用担心演示主题。由设计人员设置的每个属性的默认值在必要时被主题表覆盖。在一些特殊情况下，应用程序设计者需要覆盖用户的主题偏好。提供了一种机制来支持这些特殊情况。

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