专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

11. 发明授权

US08451278B2 Determine intended motions 有权
标题翻译：确定预期的动议
公开(公告)号：US08451278B2
公开(公告)日：2013-05-28
申请号：US13566939
申请日：2012-08-03
申请人： Kevin Geisner , Relja Markovic , Stephen Gilchrist Latta , Gregory Nelson Snook , Kudo Tsunoda , Darren Alexander Bennett
发明人： Kevin Geisner , Relja Markovic , Stephen Gilchrist Latta , Gregory Nelson Snook , Kudo Tsunoda , Darren Alexander Bennett
IPC分类号： G06T13/00
CPC分类号： A63F13/213 , A63F13/42 , A63F2300/1093 , A63F2300/6045 , A63F2300/64 , A63F2300/69 , G06F3/011 , G06F3/017 , G06T13/40
摘要： It may be desirable to apply corrective data to aspects of captured image or the user-performed gesture for display of a visual representation that corresponds to the corrective data. The captured motion may be any motion in the physical space that is captured by the capture device, such as a camera. Aspects of a skeletal or mesh model of a person, that is generated based on the image data captured by the capture device, may be modified prior to animation. The modification may be made to the model generated from image data that represents a target or a target's motion, including user gestures, in the physical space. For example, certain joints of a skeletal model may be readjusted or realigned. A model of a target may be modified by applying differential correction, magnetism principles, binary snapping, confining virtual movement to defined spaces, or the like.
摘要翻译：可能期望将校正数据应用于捕获图像的方面或用户执行的手势以显示对应于校正数据的视觉表示。捕获的运动可以是由捕获装置捕获的物理空间中的任何运动，例如相机。可以在动画之前修改基于捕获设备捕获的图像数据生成的人的骨骼或网格模型的方面。可以对在物理空间中表示目标或目标的运动（包括用户手势）的图像数据生成的模型进行修改。例如，可以重新调整或重新对准骨骼模型的某些关节。可以通过应用差分校正，磁性原理，二进制捕捉，将虚拟运动限定到定义的空间等来修改目标的模型。

12. 发明申请

US20110109617A1 Visualizing Depth 审中-公开
标题翻译：可视化深度
公开(公告)号：US20110109617A1
公开(公告)日：2011-05-12
申请号：US12617012
申请日：2009-11-12
申请人： Gregory Nelson Snook , Relja Markovic , Stephen Gilchrist Latta , Kevin Geisner
发明人： Gregory Nelson Snook , Relja Markovic , Stephen Gilchrist Latta , Kevin Geisner
IPC分类号： G06T15/00
CPC分类号： G06T17/00 , G06T19/20 , G06T2219/2012
摘要： An image such as a depth image of a scene may be received, observed, or captured by a device. The image may then be analyzed to identify one or more targets within the scene. When a target is identified, vertices may be generated. A mesh model may then be created by drawing lines that may connect the vertices. Additionally, a depth value may also be calculated for each vertex. The depth values of the vertices may then be used to extrude the mesh model such that the mesh model may represent the target in the three-dimensional virtual world. A colorization scheme, a texture, lighting effects, or the like, may be also applied to the mesh model to convey the depth the virtual object may have in the virtual world.
摘要翻译：可以由设备接收，观察或捕获诸如场景的深度图像的图像。然后可以分析图像以识别场景内的一个或多个目标。当目标被识别时，可能会产生顶点。然后可以通过绘制可以连接顶点的线来创建网格模型。另外，也可以针对每个顶点计算深度值。然后可以使用顶点的深度值来拉伸网格模型，使得网格模型可以表示三维虚拟世界中的目标。颜色化方案，纹理，照明效果等也可以应用于网格模型以传达虚拟对象可能在虚拟世界中具有的深度。

13. 发明申请

US20100238182A1 CHAINING ANIMATIONS 有权
标题翻译：链接动画
公开(公告)号：US20100238182A1
公开(公告)日：2010-09-23
申请号：US12408141
申请日：2009-03-20
申请人： Kevin Geisner , Relja Markovic , Stephen Gilchrist Latta , Gregory Nelson Snook
发明人： Kevin Geisner , Relja Markovic , Stephen Gilchrist Latta , Gregory Nelson Snook
IPC分类号： G06T13/00
CPC分类号： G06T13/80 , G06K9/00342 , G06T13/00
摘要： In applications that display a representation of a user, it may be reasonable to insert a pre-canned animation rather than animating a user's captured motion. For example, in a tennis swing, the ball toss and take back in a serve could be a pre-canned animation, whereas the actual forward swing may be mapped from the user's gestures. An animation of a user's gestures can be chained together into sequences with pre-canned animations, where animation blending techniques can provide for a smoother transition between the animation types. Techniques for blending animations, that may comprise determining boundaries and transition points between pre-canned animations and animations based on captured motion, may improve animation efficiency. Gesture history, including joint position, velocity, and acceleration, can be used to determine user intent, seed parameters for subsequent animations and game control, and determine the subsequent gestures to initiate.
摘要翻译：在显示用户表示的应用程序中，插入预先准备的动画可能是合理的，而不是动画化用户的捕获动作。例如，在网球摆动中，球投掷并收回服务可以是预先制作的动画，而实际的向前摆动可以从用户的手势映射。用户的手势的动画可以一起链接到具有预先固定动画的序列中，其中动画混合技术可以提供动画类型之间更平滑的转换。用于混合动画的技术，其可以包括基于捕获的运动来确定预先确定的动画和动画之间的边界和转换点，可以提高动画效率。手势历史，包括联合位置，速度和加速度，可用于确定用户意图，后续动画和游戏控制的种子参数，并确定随后的手势启动。

14. 发明授权

US09041623B2 Total field of view classification for head-mounted display 有权
标题翻译：头戴式显示器的总体视野分类
公开(公告)号：US09041623B2
公开(公告)日：2015-05-26
申请号：US13692885
申请日：2012-12-03
申请人： James Chia-Ming Liu , Stephen Gilchrist Latta , Anton Andrews , Benjamin Isaac Vaught , Christopher Michael Novak , Sheridan Leigh Martin
发明人： James Chia-Ming Liu , Stephen Gilchrist Latta , Anton Andrews , Benjamin Isaac Vaught , Christopher Michael Novak , Sheridan Leigh Martin
IPC分类号： G09G5/377 , G02B27/01
CPC分类号： G09G5/38 , G02B27/0093 , G02B27/017 , G02B27/0172 , G02B2027/0187 , G06F3/011 , G06F3/017 , G09G5/377
摘要： Virtual images are located for display in a head-mounted display (HMD) to provide an augment reality view to an HMD wearer. Sensor data may be collected from on-board sensors provided on an HMD. Additionally, other day may be collected from external sources. Based on the collected sensor data and other data, the position and rotation of the HMD wearer's head relative to the HMD wearer's body and surrounding environment may be determined. After resolving the HMD wearer's head position, the HMD wearer's total field of view (TFOV) may be classified into regions. Virtual images may then be located in the classified TFOV regions to locate the virtual images relative to the HMD wearer's body and surrounding environment.
摘要翻译：虚拟图像位于用于显示在头戴式显示器（HMD）中以向HMD佩戴者提供增强现实视图。可以从HMD上提供的车载传感器收集传感器数据。另外，另外一天可能从外部来源收集。基于所收集的传感器数据和其他数据，可以确定HMD佩戴者的头部相对于HMD佩戴者身体和周围环境的位置和旋转。在解决HMD佩戴者的头部位置之后，HMD佩戴者的总视野（TFOV）可以分为区域。然后虚拟图像可以位于分类的TFOV区域中，以相对于HMD佩戴者的身体和周围环境定位虚拟图像。

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