公开(公告)号：US07668340B2

公开(公告)日：2010-02-23

申请号：US12326540

申请日：2008-12-02

申请人： Charles J. Cohen ,  Glenn J. Beach ,  Brook Cavell ,  Eugene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George V. Paul

发明人： Charles J. Cohen ,  Glenn J. Beach ,  Brook Cavell ,  Eugene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George V. Paul

IPC分类号： G06K9/00 ,  G06F3/033

CPC分类号： G06K9/00355 ,  A63F2300/1093 ,  A63F2300/6045 ,  G06F3/017 ,  G06T7/246

摘要： A gesture recognition interface for use in controlling self-service machines and other devices is disclosed. A gesture is defined as motions and kinematic poses generated by humans, animals, or machines. Specific body features are tracked, and static and motion gestures are interpreted. Motion gestures are defined as a family of parametrically delimited oscillatory motions, modeled as a linear-in-parameters dynamic system with added geometric constraints to allow for real-time recognition using a small amount of memory and processing time. A linear least squares method is preferably used to determine the parameters which represent each gesture. Feature position measure is used in conjunction with a bank of predictor bins seeded with the gesture parameters, and the system determines which bin best fits the observed motion. Recognizing static pose gestures is preferably performed by localizing the body/object from the rest of the image, describing that object, and identifying that description. The disclosure details methods for gesture recognition, as well as the overall architecture for using gesture recognition to control of devices, including self-service machines.

摘要翻译： 公开了一种用于控制自助服务机器和其他设备的手势识别界面。 手势被定义为人类，动物或机器产生的运动和运动姿态。 跟踪特定身体特征，并解释静态和运动手势。 运动手势被定义为参数定界的振荡运动系列，被建模为参数线性参数动态系统，附加的几何约束允许使用少量的存储器和处理时间进行实时识别。 优选使用线性最小二乘法来确定表示每个姿态的参数。 特征位置测量结合使用手势参数种子的预测器仓组，并且系统确定哪个箱最适合观察到的运动。 识别静态姿势手势优选地通过将来自图像的其余部分的身体/物体定位，描述该对象并识别该描述来执行。 本公开详细描述了用于手势识别的方法，以及用于使用手势识别来控制设备（包括自助服务机器）的整体架构。

公开(公告)号：US07036094B1

公开(公告)日：2006-04-25

申请号：US09540461

申请日：2000-03-31

申请人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

发明人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

IPC分类号： G09G5/00

CPC分类号： G06F3/017 ,  G06K9/00348 ,  G06K9/00355

摘要： A system for recognizing various human and creature motion gaits and behaviors is presented. These behaviors are defined as combinations of “gestures” identified on various parts of a body in motion. For example, the leg gestures generated when a person runs are different than when a person walks. The system described here can identify such differences and categorize these behaviors. Gestures, as previously defined, are motions generated by humans, animals, or machines. Where in the previous patent only one gesture was recognized at a time, in this system, multiple gestures on a body (or bodies) are recognized simultaneously and used in determining behaviors. If multiple bodies are tracked by the system, then overall formations and behaviors (such as military goals) can be determined.

摘要翻译： 提出了一种用于识别各种人类和生物运动步态和行为的系统。 这些行为被定义为在运动的身体的各个部分上识别的“手势”的组合。 例如，当人运行时产生的腿姿势与人走路时不同。 这里描述的系统可以识别这些差异并对这些行为进行分类。 如前所述，手势是由人，动物或机器产生的动作。 在以前的专利中，一次只能识别出一个手势，在该系统中，身体（或身体）上的多个手势被同时识别并用于确定行为。 如果系统跟踪多个物体，则可以确定整体形成和行为（如军事目标）。

公开(公告)号：US06950534B2

公开(公告)日：2005-09-27

申请号：US10759459

申请日：2004-01-16

申请人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

发明人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

IPC分类号： G06F3/00 ,  G06F3/01 ,  G06K9/00 ,  G06T7/20

CPC分类号： G06K9/00355 ,  A63F2300/1093 ,  A63F2300/6045 ,  G06F3/017 ,  G06T7/246

摘要： A gesture recognition interface for use in controlling self-service machines and other devices is disclosed. A gesture is defined as motions and kinematic poses generated by humans, animals, or machines. Specific body features are tracked, and static and motion gestures are interpreted. Motion gestures are defined as a family of parametrically delimited oscillatory motions, modeled as a linear-in-parameters dynamic system with added geometric constraints to allow for real-time recognition using a small amount of memory and processing time. A linear least squares method is preferably used to determine the parameters which represent each gesture. Feature position measure is used in conjunction with a bank of predictor bins seeded with the gesture parameters, and the system determines which bin best fits the observed motion. Recognizing static pose gestures is preferably performed by localizing the body/object from the rest of the image, describing that object, and identifying that description. The disclosure details methods for gesture recognition, as well as the overall architecture for using gesture recognition to control of devices, including self-service machines.

摘要翻译： 公开了一种用于控制自助服务机器和其他设备的手势识别界面。 手势被定义为人类，动物或机器产生的运动和运动姿态。 跟踪特定身体特征，并解释静态和运动手势。 运动手势被定义为参数定界的振荡运动系列，被建模为参数线性参数动态系统，附加的几何约束允许使用少量的存储器和处理时间进行实时识别。 优选使用线性最小二乘法来确定表示每个姿态的参数。 特征位置测量结合使用手势参数种子的预测器仓组，并且系统确定哪个箱最适合观察到的运动。 识别静态姿势手势优选地通过将来自图像的其余部分的身体/物体定位，描述该对象并识别该描述来执行。 本公开详细描述了用于手势识别的方法，以及用于使用手势识别来控制设备（包括自助服务机器）的整体架构。

公开(公告)号：US08407625B2

公开(公告)日：2013-03-26

申请号：US11412252

申请日：2006-04-25

申请人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

发明人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

IPC分类号： G06F3/048

CPC分类号： G06F3/017 ,  G06K9/00348 ,  G06K9/00355

摘要： A system for recognizing various human and creature motion gaits and behaviors is presented. These behaviors are defined as combinations of “gestures” identified on various parts of a body in motion. For example, the leg gestures generated when a person runs are different than when a person walks. The system described here can identify such differences and categorize these behaviors. Gestures, as previously defined, are motions generated by humans, animals, or machines. Multiple gestures on a body (or bodies) are recognized simultaneously and used in determining behaviors. If multiple bodies are tracked by the system, then overall formations and behaviors (such as military goals) can be determined.

摘要翻译： 提出了一种用于识别各种人类和生物运动步态和行为的系统。 这些行为被定义为在运动的身体的各个部分上识别的手势的组合。 例如，当人运行时产生的腿姿势与人走路时不同。 这里描述的系统可以识别这些差异并对这些行为进行分类。 如前所述，手势是由人，动物或机器产生的动作。 身体（或身体）上的多个手势被同时识别并用于确定行为。 如果系统跟踪多个物体，则可以确定整体形成和行为（如军事目标）。

公开(公告)号：US20090074248A1

公开(公告)日：2009-03-19

申请号：US12326540

申请日：2008-12-02

申请人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

发明人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

IPC分类号： G06K9/00

CPC分类号： G06K9/00355 ,  A63F2300/1093 ,  A63F2300/6045 ,  G06F3/017 ,  G06T7/246

摘要： A gesture recognition interface for use in controlling self-service machines and other devices is disclosed. A gesture is defined as motions and kinematic poses generated by humans, animals, or machines. Specific body features are tracked, and static and motion gestures are interpreted. Motion gestures are defined as a family of parametrically delimited oscillatory motions, modeled as a linear-in-parameters dynamic system with added geometric constraints to allow for real-time recognition using a small amount of memory and processing time. A linear least squares method is preferably used to determine the parameters which represent each gesture. Feature position measure is used in conjunction with a bank of predictor bins seeded with the gesture parameters, and the system determines which bin best fits the observed motion. Recognizing static pose gestures is preferably performed by localizing the body/object from the rest of the image, describing that object, and identifying that description. The disclosure details methods for gesture recognition, as well as the overall architecture for using gesture recognition to control of devices, including self-service machines.

摘要翻译： 公开了一种用于控制自助服务机器和其他设备的手势识别界面。 手势被定义为人类，动物或机器产生的运动和运动姿态。 跟踪特定身体特征，并解释静态和运动手势。 运动手势被定义为参数定界的振荡运动系列，被建模为参数线性参数动态系统，附加的几何约束允许使用少量的存储器和处理时间进行实时识别。 优选使用线性最小二乘法来确定表示每个姿态的参数。 特征位置测量结合使用手势参数种子的预测器仓组，并且系统确定哪个箱最适合观察到的运动。 识别静态姿势手势优选地通过将来自图像的其余部分的身体/物体定位，描述该对象并识别该描述来执行。 本公开详细描述了用于手势识别的方法，以及用于使用手势识别来控制设备（包括自助服务机器）的整体架构。

公开(公告)号：US06681031B2

公开(公告)日：2004-01-20

申请号：US09371460

申请日：1999-08-10

申请人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

发明人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

IPC分类号： G06K900

CPC分类号： G06F3/017 ,  A63F2300/1093 ,  A63F2300/6045 ,  G06K9/00355 ,  G06T7/246 ,  G07F9/023

摘要： A gesture recognition interface for use in controlling self-service machines and other devices is disclosed. A gesture is defined as motions and kinematic poses generated by humans, animals, or machines. Specific body features are tracked, and static and motion gestures are interpreted. Motion gestures are defined as a family of parametrically delimited oscillatory motions, modeled as a linear-in-parameters dynamic system with added geometric constraints to allow for real-time recognition using a small amount of memory and processing time. A linear least squares method is preferably used to determine the parameters which represent each gesture. Feature position measure is used in conjunction with a bank of predictor bins seeded with the gesture parameters, and the system determines which bin best fits the observed motion. Recognizing static pose gestures is preferably performed by localizing the body/object from the rest of the image, describing that object, and identifying that description. The disclosure details methods for gesture recognition, as well as the overall architecture for using gesture recognition to control of devices, including self-service machines.

摘要翻译： 公开了一种用于控制自助服务机器和其他设备的手势识别界面。 手势被定义为人类，动物或机器产生的运动和运动姿态。 跟踪特定身体特征，并解释静态和运动手势。 运动手势被定义为参数定界的振荡运动系列，被建模为参数线性参数动态系统，附加的几何约束允许使用少量的存储器和处理时间进行实时识别。 优选使用线性最小二乘法来确定表示每个姿态的参数。 特征位置测量结合使用手势参数种子的预测器仓组，并且系统确定哪个箱最适合观察到的运动。 识别静态姿势手势优选地通过将来自图像的其余部分的身体/物体定位，描述该对象并识别该描述来执行。 本公开详细描述了用于手势识别的方法，以及用于使用手势识别来控制设备（包括自助服务机器）的整体架构。

公开(公告)号：US07460690B2

公开(公告)日：2008-12-02

申请号：US11226831

申请日：2005-09-14

申请人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

发明人： Charles J. Cohen ,  Glenn Beach ,  Brook Cavell ,  Gene Foulk ,  Charles J. Jacobus ,  Jay Obermark ,  George Paul

IPC分类号： G06K9/00 ,  G06K9/36 ,  G06F3/033

CPC分类号： G06K9/00355 ,  A63F2300/1093 ,  A63F2300/6045 ,  G06F3/017 ,  G06T7/246

摘要： A gesture recognition interface for use in controlling self-service machines and other devices is disclosed. A gesture is defined as motions and kinematic poses generated by humans, animals, or machines. Specific body features are tracked, and static and motion gestures are interpreted. Motion gestures are defined as a family of parametrically delimited oscillatory motions, modeled as a linear-in-parameters dynamic system with added geometric constraints to allow for real-time recognition using a small amount of memory and processing time. A linear least squares method is preferably used to determine the parameters which represent each gesture. Feature position measure is used in conjunction with a bank of predictor bins seeded with the gesture parameters, and the system determines which bin best fits the observed motion. Recognizing static pose gestures is preferably performed by localizing the body/object from the rest of the image, describing that object, and identifying that description. The disclosure details methods for gesture recognition, as well as the overall architecture for using gesture recognition to control of devices, including self-service machines.

摘要翻译： 公开了一种用于控制自助服务机器和其他设备的手势识别界面。 手势被定义为人类，动物或机器产生的运动和运动姿态。 跟踪特定身体特征，并解释静态和运动手势。 运动手势被定义为参数定界的振荡运动系列，被建模为参数线性参数动态系统，附加的几何约束允许使用少量的存储器和处理时间进行实时识别。 优选使用线性最小二乘法来确定表示每个姿态的参数。 特征位置测量结合使用手势参数种子的预测器仓组，并且系统确定哪个箱最适合观察到的运动。 识别静态姿势手势优选地通过将来自图像的其余部分的身体/物体定位，描述该对象并识别该描述来执行。 本公开详细描述了用于手势识别的方法，以及用于使用手势识别来控制设备（包括自助服务机器）的整体架构。

公开(公告)号：US20090116692A1

公开(公告)日：2009-05-07

申请号：US12013717

申请日：2008-01-14

申请人： George V. Paul ,  Glenn J. Beach ,  Charles J. Cohen ,  Charles J. Jacobus

发明人： George V. Paul ,  Glenn J. Beach ,  Charles J. Cohen ,  Charles J. Jacobus

IPC分类号： G06K9/00

CPC分类号： G06K9/00335 ,  A63F2300/1093 ,  A63F2300/6045 ,  A63F2300/69 ,  G06F3/017 ,  G06K9/00355 ,  G06T7/246

摘要： A real-time computer vision system tracks one or more objects moving in a scene using a target location technique which does not involve searching. The imaging hardware includes a color camera, frame grabber and processor. The software consists of the low-level image grabbing software and a tracking algorithm. The system tracks objects based on the color, motion and/or shape of the object in the image. A color matching function is used to compute three measures of the target's probable location based on the target color, shape and motion. The method then computes the most probable location of the target using a weighting technique. Once the system is running, a graphical user interface displays the live image from the color camera on the computer screen. The operator can then use the mouse to select a target for tracking. The system will then keep track of the moving target in the scene in real-time.

摘要翻译： 实时计算机视觉系统使用不涉及搜索的目标定位技术跟踪在场景中移动的一个或多个对象。 成像硬件包括彩色摄像机，帧采集器和处理器。 该软件由低级图像抓取软件和跟踪算法组成。 该系统基于图像中对象的颜色，运动和/或形状来跟踪对象。 颜色匹配功能用于基于目标颜色，形状和运动计算目标可能位置的三个度量。 该方法然后使用加权技术计算目标的最可能的位置。 一旦系统运行，图形用户界面将在计算机屏幕上显示来自彩色摄像机的实时图像。 然后，操作员可以使用鼠标选择要跟踪的目标。 然后系统将实时跟踪场景中的移动目标。

公开(公告)号：US07050606B2

公开(公告)日：2006-05-23

申请号：US10004058

申请日：2001-11-01

申请人： George V. Paul ,  Glenn J. Beach ,  Charles J. Cohen ,  Charles J. Jacobus

发明人： George V. Paul ,  Glenn J. Beach ,  Charles J. Cohen ,  Charles J. Jacobus

IPC分类号： G06K9/00

CPC分类号： G06K9/00355 ,  A63F2300/1093 ,  A63F2300/6045 ,  A63F2300/69 ,  G06F3/017 ,  G06K9/00335 ,  G06K9/4652 ,  G06K9/6212 ,  G06T7/246 ,  G06T7/251 ,  G06T2207/10016 ,  G06T2207/30196 ,  G06T2207/30268

摘要： A system and method tracks the movements of a driver or passenger in a vehicle (ground, water, air, or other) and controls devices in accordance with position, motion, and/or body or hand gestures or movements. According to one embodiment, an operator or passenger uses the invention to control comfort or entertainment features such the heater, air conditioner, lights, mirror positions or the radio/CD player using hand gestures. An alternative embodiment facilitates the automatic adjustment of car seating restraints based on head position. Yet another embodiment is used to determine when to fire an airbag (and at what velocity or orientation) based on the position of a person in a vehicle seat. The invention may also be used to control systems outside of the vehicle. The on-board sensor system would be used to track the driver or passenger, but when the algorithms produce a command for a desired response, that response (or just position and gesture information) could be transmitted via various methods (wireless, light, whatever) to other systems outside the vehicle to control devices located outside the vehicle. For example, this would allow a person to use gestures inside the car to interact with a kiosk located outside of the car.

公开(公告)号：US07684592B2

公开(公告)日：2010-03-23

申请号：US12013717

申请日：2008-01-14

申请人： George V. Paul ,  Glenn J. Beach ,  Charles J. Cohen ,  Charles J. Jacobus

发明人： George V. Paul ,  Glenn J. Beach ,  Charles J. Cohen ,  Charles J. Jacobus

IPC分类号： G06K9/00

CPC分类号： G06K9/00335 ,  A63F2300/1093 ,  A63F2300/6045 ,  A63F2300/69 ,  G06F3/017 ,  G06K9/00355 ,  G06T7/246

摘要： A real-time computer vision system tracks one or more objects moving in a scene using a target location technique which does not involve searching. The imaging hardware includes a color camera, frame grabber and processor. The software consists of the low-level image grabbing software and a tracking algorithm. The system tracks objects based on the color, motion and/or shape of the object in the image. A color matching function is used to compute three measures of the target's probable location based on the target color, shape and motion. The method then computes the most probable location of the target using a weighting technique. Once the system is running, a graphical user interface displays the live image from the color camera on the computer screen. The operator can then use the mouse to select a target for tracking. The system will then keep track of the moving target in the scene in real-time.

摘要翻译： 实时计算机视觉系统使用不涉及搜索的目标定位技术跟踪在场景中移动的一个或多个对象。 成像硬件包括彩色摄像机，帧采集器和处理器。 该软件由低级图像抓取软件和跟踪算法组成。 该系统基于图像中对象的颜色，运动和/或形状来跟踪对象。 颜色匹配功能用于基于目标颜色，形状和运动计算目标可能位置的三个度量。 该方法然后使用加权技术计算目标的最可能的位置。 一旦系统运行，图形用户界面将在计算机屏幕上显示来自彩色摄像机的实时图像。 然后，操作员可以使用鼠标选择要跟踪的目标。 然后系统将实时跟踪场景中的移动目标。