公开(公告)号：US20150312523A1

公开(公告)日：2015-10-29

申请号：US13997266

申请日：2012-04-09

申请人： WENLONG LI ,  XIAOFENG TONG ,  YANGZHOU DU ,  WEI HU ,  YIMIN ZHANG

发明人： WENLONG LI ,  XIAOFENG TONG ,  YANGZHOU DU ,  WEI HU ,  YIMIN ZHANG

IPC分类号： H04N7/15 ,  G06F3/0484 ,  G06F3/0488 ,  G06K9/00

CPC分类号： H04N7/157 ,  G06F3/04842 ,  G06F3/0488 ,  G06K9/00255 ,  H04M1/72544 ,  H04M2250/22 ,  H04M2250/52

摘要： A video communication system that replaces actual live images of the participating users with animated avatars. A method may include selecting an avatar, initiating communication, capturing an image, detecting a face in the image, extracting features from the face, converting the facial features to avatar parameters, and transmitting at least one of the avatar selection or avatar parameters. The system is configured to allow a user to select a new avatar during active communication with a remote user.

摘要翻译： 视频通信系统，用动画头像代替参与用户的实际实况图像。 方法可以包括选择化身，发起通信，拍摄图像，检测图像中的脸部，从脸部提取特征，将面部特征转换成化身参数，以及发送化身选择或化身参数中的至少一个。 该系统被配置为允许用户在与远程用户的主动通信期间选择新的化身。

公开(公告)号：US20140300539A1

公开(公告)日：2014-10-09

申请号：US14271191

申请日：2014-05-06

申请人： Xiaofeng Tong ,  Dayong Ding ,  Wenlong Li ,  Yimin Zhang

发明人： Xiaofeng Tong ,  Dayong Ding ,  Wenlong Li ,  Yimin Zhang

IPC分类号： G06K9/00 ,  G06F3/01

CPC分类号： G06K9/00355 ,  G06F3/017 ,  G06K9/00228 ,  G06K9/00261

摘要： Methods, apparatuses, and articles associated with gesture recognition using depth images are disclosed herein. In various embodiments, an apparatus may include a face detection engine configured to determine whether a face is present in one or more gray images of respective image frames generated by a depth camera, and a hand tracking engine configured to track a hand in one or more depth images generated by the depth camera. The apparatus may further include a feature extraction and gesture inference engine configured to extract features based on results of the tracking by the hand tracking engine, and infer a hand gesture based at least in part on the extracted features. Other embodiments may also be disclosed and claimed.

公开(公告)号：US20130332834A1

公开(公告)日：2013-12-12

申请号：US13995484

申请日：2011-09-12

申请人： Wenlong Li ,  Yangzhou Du ,  Xiaofeng Tong ,  Yimin Zhang

发明人： Wenlong Li ,  Yangzhou Du ,  Xiaofeng Tong ,  Yimin Zhang

IPC分类号： G06F3/0484

CPC分类号： G06F3/0484 ,  G06F16/78

摘要： Methods, apparatuses and storage medium associated with cooperative annotation and/or recommendation by shared and personal devices. In various embodiments, at least one non-transitory computer-readable storage medium may include a number of instructions configured to enable a personal device (PD) of a user, in response to execution of the instructions by the personal device, to receive a user input selecting performance of a user function in association with a video stream being rendered on a shared video device (SVD) configured for use by multiple users, render an image frame of the video stream rendered on the shared video device at a time proximate to a time of the user input, and facilitate performance of the user function, which may include annotation of video objects. Other embodiments, including recommendation of video content, may be disclosed or claimed.

摘要翻译： 与共同和个人设备的协作注释和/或推荐相关联的方法，设备和存储介质。 在各种实施例中，至少一个非暂时计算机可读存储介质可以包括多个指令，其被配置为响应于个人设备的指令的执行而使用户的个人设备（PD）能够接收用户 与被配置为由多个用户使用的共享视频设备（SVD）上呈现的视频流相关联的用户功能的输入选择性能，在接近于一个时间的时刻渲染在共享视频设备上呈现的视频流的图像帧 用户输入的时间，并且便于用户功能的执行，其可以包括视频对象的注释。 可以公开或要求保护包括视频内容的推荐的其他实施例。

公开(公告)号：US20130297650A1

公开(公告)日：2013-11-07

申请号：US13994768

申请日：2011-09-12

申请人： Wenlong Li ,  Xiaofeng Tong ,  Yimin Zhang

发明人： Wenlong Li ,  Xiaofeng Tong ,  Yimin Zhang

IPC分类号： G06F17/30

CPC分类号： G06F16/43 ,  G06F16/434

摘要： A product in television program currently being watched can be identified by extracting at least one decoded frame from a television transmission. The frame can be transmitted to a separate mobile device for requesting an image search and for receiving the search results. The search results can be used to identify the product.

摘要翻译： 目前正在观看的电视节目中的产品可以通过从电视传输中提取至少一个解码的帧来识别。 该帧可以被发送到单独的移动设备，用于请求图像搜索和接收搜索结果。 搜索结果可用于识别产品。

公开(公告)号：US20130278504A1

公开(公告)日：2013-10-24

申请号：US13977248

申请日：2011-11-01

申请人： Xiaofeng Tong ,  Dayong Ding ,  Wenlong Li

发明人： Xiaofeng Tong ,  Dayong Ding ,  Wenlong Li

IPC分类号： G06F3/01

CPC分类号： G06F3/017 ,  G06K9/00355

摘要： Systems, devices and methods are described including starting a gesture recognition engine in response to detecting an initiation gesture and using the gesture recognition engine to determine a hand posture and a hand trajectory in various depth images. The gesture recognition engine may then use the hand posture and the hand trajectory to recognize a dynamic hand gesture and provide corresponding user interface command.

摘要翻译： 描述了系统，设备和方法，包括响应于检测到启动手势而启动手势识别引擎，并使用手势识别引擎确定各种深度图像中的手势和手轨迹。 手势识别引擎然后可以使用手势和手迹来识别动态手势并提供对应的用户界面命令。

公开(公告)号：US08553982B2

公开(公告)日：2013-10-08

申请号：US12646807

申请日：2009-12-23

申请人： Xiaofeng Tong ,  Wenlong Li ,  Yimin Zhang

发明人： Xiaofeng Tong ,  Wenlong Li ,  Yimin Zhang

IPC分类号： G06K9/46 ,  G06K9/00 ,  G06K9/34 ,  G06K9/50 ,  G06K9/52 ,  H04N7/18 ,  H04N5/225

CPC分类号： G06T3/0068 ,  G06T7/35

摘要： A method, apparatus, and system are described for model-based playfield registration. An input video image is processed. The processing of the video image includes extracting key points relating to the video image. Further, whether enough key points relating to the video image were extracted is determined, and a direct estimation of the video image is performed if enough key points have been extracted and then, a homograph matrix of a final video image based on the direct estimation is generated.

摘要翻译： 描述了基于模型的游戏场注册的方法，装置和系统。 输入视频图像被处理。 视频图像的处理包括提取与视频图像相关的关键点。 此外，是否提取了与视频图像有关的足够的关键点，并且如果已经提取了足够的关键点，则执行视频图像的直接估计，然后基于直接估计的最终视频图像的同形矩阵是 生成。

公开(公告)号：US20110150275A1

公开(公告)日：2011-06-23

申请号：US12646807

申请日：2009-12-23

申请人： Xiaofeng Tong ,  Wenlong Li ,  Yimin Zhang

发明人： Xiaofeng Tong ,  Wenlong Li ,  Yimin Zhang

IPC分类号： G06T7/00

CPC分类号： G06T3/0068 ,  G06T7/35

摘要： A method, apparatus, and system are described for model-based playfield registration. An input video image is processed. The processing of the video image includes extracting key points relating to the video image. Further, whether enough key points relating to the video image were extracted is determined, and a direct estimation of the video image is performed if enough key points have been extracted and then, a homograph matrix of a final video image based on the direct estimation is generated.

摘要翻译： 描述了基于模型的游戏场注册的方法，装置和系统。 输入视频图像被处理。 视频图像的处理包括提取与视频图像相关的关键点。 此外，是否提取了与视频图像相关的足够的关键点，并且如果已经提取了足够的关键点，则执行视频图像的直接估计，然后基于直接估计的最终视频图像的同形矩阵是 生成。

公开(公告)号：US20100329560A1

公开(公告)日：2010-12-30

申请号：US12459141

申请日：2009-06-26

申请人： Changcheng Li ,  XiaoFeng Tong

发明人： Changcheng Li ,  XiaoFeng Tong

IPC分类号： G06K9/34 ,  G06K9/46

CPC分类号： G06K9/00342 ,  G06K9/6297

摘要： The present invention discloses a method of estimating human pose comprising: modeling a human body as a tree structure; optimizing said tree structure through importance proposal probabilities and part priorities; performing foreground detection to create image region observation; and performing image segmentation to provide image edge observations.

摘要翻译： 本发明公开了一种估计人体姿态的方法，包括：将人体建模为树状结构; 通过重要性提案概率和部分优先级优化树结构; 执行前景检测以创建图像区域观察; 并执行图像分割以提供图像边缘观察。

公开(公告)号：US20090147992A1

公开(公告)日：2009-06-11

申请号：US11953818

申请日：2007-12-10

申请人： Xiaofeng Tong ,  Yimin Zhang

发明人： Xiaofeng Tong ,  Yimin Zhang

IPC分类号： G06K9/00

CPC分类号： G06K9/00335 ,  A63B2024/0056 ,  G06T7/251 ,  G06T2207/10016 ,  G06T2207/20061 ,  G06T2207/30224 ,  G06T2207/30241

摘要： A three-level ball detection and tracking method is disclosed. The ball detection and tracking method employs three levels to generate multiple ball candidates rather than a single one. The ball detection and tracking method constructs multiple trajectories using candidate linking, then uses optimization criteria to determine the best ball trajectory.

摘要翻译： 公开了一种三级球检测和跟踪方法。 球检测和跟踪方法采用三个级别来生成多个候选球，而不是单个球。 球检测和跟踪方法使用候选链接构建多个轨迹，然后使用优化标准来确定最佳球轨迹。