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

US20090228251A1 Systems and Methods for Designing Optical Surfaces 审中-公开
标题翻译：用于设计光学表面的系统和方法
公开(公告)号：US20090228251A1
公开(公告)日：2009-09-10
申请号：US12267992
申请日：2008-11-10
申请人： Ozan Cakmakci , Jannick Rolland , Brendan Moore , Hassan Foroosh
发明人： Ozan Cakmakci , Jannick Rolland , Brendan Moore , Hassan Foroosh
IPC分类号： G06F17/10
CPC分类号： G06T17/30
摘要： In one embodiment, a system and method for designing an optical surface pertain to constructing a matrix of radial basis functions, each column of the matrix corresponding to a radial basis function, selecting an initial surface to be used as a starting point, selecting initial weights for the radial basis functions, generating an initialized optical surface based upon the selected initial surface and initial weights, and optimizing the initialized optical surface to minimize an error function.
摘要翻译：在一个实施例中，用于设计光学表面的系统和方法涉及构造径向基函数的矩阵，矩阵的每列对应于径向基函数，选择要用作起始点的初始表面，选择初始权重对于径向基函数，基于所选择的初始表面和初始权重生成初始化的光学表面，以及优化初始化的光学表面以最小化误差函数。

2. 发明授权

US08587686B1 Hybrid differential optical sensing imager 失效
标题翻译：混合差分光学传感成像仪
公开(公告)号：US08587686B1
公开(公告)日：2013-11-19
申请号：US13049478
申请日：2011-03-16
申请人： Nabeel A. Riza , Hassan Foroosh
发明人： Nabeel A. Riza , Hassan Foroosh
IPC分类号： H04N5/235 , H04N5/228
CPC分类号： G01S17/42 , G01S7/4802 , G01S7/481 , G01S17/023 , G01S17/89
摘要： Methods and systems for a differential compressed sensor to form a smart optical imager that uses both active (laser) and passive (ambient light) to sample the direct image information within a three-dimensional spatial frame. The lens-based distance sensor is produces smart sampling of target by adjusting the size of the laser beam spot on the target sampling grid to produce a boundary outline by a light flooding. This target dependent direct sampling of the target results in direct compressed sensing. A passive light acquisition pin-hole sampling optical sensor design is proposed that produces the pixel-basis Laplacian to determine the compressed sensed pixels in the incident image.
摘要翻译：用于差分压缩传感器以形成智能光学成像器的方法和系统，其使用有源（激光）和被动（环境光）来在三维空间帧内采样直接图像信息。基于镜头的距离传感器通过调整目标采样网格上的激光束斑点的尺寸，通过光驱产生边界轮廓，产生目标的智能采样。目标依赖的目标直接采样导致直接压缩感测。提出了一种无源光采集针孔采样光学传感器设计，其产生基于像素的拉普拉斯算子来确定入射图像中的压缩感测像素。

3. 发明授权

US08755569B2 Methods for recognizing pose and action of articulated objects with collection of planes in motion 有权
标题翻译：用于识别运动中收集飞机的铰接物体的姿势和动作的方法
公开(公告)号：US08755569B2
公开(公告)日：2014-06-17
申请号：US12790083
申请日：2010-05-28
申请人： Yuping Shen , Hassan Foroosh
发明人： Yuping Shen , Hassan Foroosh
IPC分类号： G06K9/00
CPC分类号： G06K9/00335
摘要： The invention comprises an improved system, method, and computer-readable instructions for recognizing pose and action of articulated objects with collection of planes in motion. The method starts with a video sequence and a database of reference sequences corresponding to different known actions. The method identifies the sequence from the reference sequences such that the subject in performs the closest action to that observed. The method compares actions by comparing pose transitions. The cross-homography invariant may be used for view-invariant recognition of human body pose transition and actions.
摘要翻译：本发明包括改进的系统，方法和计算机可读指令，用于通过收集运动中的平面来识别铰接对象的姿态和动作。该方法以视频序列和对应于不同已知动作的参考序列的数据库开始。该方法从参考序列识别序列，使得受试者对观察到的序列进行最接近的作用。该方法通过比较姿态转换来比较动作。交叉单应性不变量可用于视觉不变的人体姿态转换和动作识别。

4. 发明申请

US20100303303A1 METHODS FOR RECOGNIZING POSE AND ACTION OF ARTICULATED OBJECTS WITH COLLECTION OF PLANES IN MOTION 有权
标题翻译：用于识别运动中采集运动物体的姿势和动作的方法
公开(公告)号：US20100303303A1
公开(公告)日：2010-12-02
申请号：US12790083
申请日：2010-05-28
申请人： Yuping Shen , Hassan Foroosh
发明人： Yuping Shen , Hassan Foroosh
IPC分类号： G06K9/00
CPC分类号： G06K9/00335
摘要： The invention comprises an improved system, method, and computer-readable instructions for recognizing pose and action of articulated objects with collection of planes in motion. The method starts with a video sequence and a database of reference sequences corresponding to different known actions. The method identifies the sequence from the reference sequences such that the subject in performs the closest action to that observed. The method compares actions by comparing pose transitions. The cross-homography invariant may be used for view-invariant recognition of human body pose transition and actions.
摘要翻译：本发明包括改进的系统，方法和计算机可读指令，用于通过收集运动中的平面来识别铰接对象的姿态和动作。该方法以视频序列和对应于不同已知动作的参考序列的数据库开始。该方法从参考序列识别序列，使得受试者对观察到的序列进行最接近的作用。该方法通过比较姿态转换来比较动作。交叉单应性不变量可用于视觉不变的人体姿态转换和动作识别。

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