Abstract: A computer-implemented method and system for controlling various electronic devices by recognition of gestures made by a user within a particular space defined in front of the user are provided. An example method may comprise generating a depth map of a physical scene, determining that a head of the user is directed towards a predetermined direction, establishing a virtual sensing zone defined between the user and a predetermined location, identifying a particular gesture made by the user within the virtual sensing zone, and selectively providing to the electronic device a control command associated with the particular gesture. The particular gesture may be performed by one or more characteristic forms provided by the user within the virtual sensing zone being in an active state. The characteristic forms are forms reliably distinguishable from casual forms by means of computer vision and having certain attributes, which can reliably reflect user intent.

Abstract: Provided are computer-implemented methods and systems for controlling devices using gestures and a 3D sensor that enables implementing the above. In one embodiment, the method proposed herein may be based on defining at least one sensor area within the space surrounding a user of a controlled device; associating this sensor area with at least one user gesture; associating the combination of the user gesture and sensor area with an actionable command; identifying the direction of the line of sight of the user and a focal point of the line of sight of the user; and, if the line of sight of the user is directed a sensor area, issuing an actionable command corresponding to the combination of the sensor area and the gesture that the user makes while looking at this sensor area.

Abstract: The present technology refers to methods for dynamic determining location and orientation of handheld device, such as a smart phone, remote controller or gaming device, within a 3D environment in real time. For these ends, there is provided a 3D camera for capturing a depth map of the 3D environment within which there is a user holding the handheld device. The handheld device acquires motion and orientation data in response to hand gestures, which data is further processed and associated with a common coordinate system. The depth map is also processed to generate motion data of user hands, which is then dynamically compared to the processed motion and orientation data obtained from the handheld device so as to determine the handheld device location and orientation. The positional and orientation data may be further used in various software applications to generate control commands or perform analysis of various gesture motions.

Abstract: Disclosed are methods for determining and tracking a current location of a handheld pointing device, such as a remote control for an entertainment system, on a depth map generated by a gesture recognition control system. The methods disclosed herein enable identifying a user's hand gesture, and generating corresponding motion data. Further, the handheld pointing device may send motion, such as acceleration or velocity, and/or orientation data such as pitch, roll, and yaw angles. The motion data of user's hand gesture and motion data (orientation data) as received from the handheld pointing device are then compared, and if they correspond to each other, it is determined that the handheld pointing device is in active use by the user as it is held by a particular hand. Accordingly, a location of the handheld pointing device on the depth map can be determined.

Abstract: A computer-implemented method for controlling one or more electronic devices by recognition of gestures made by a three-dimensional object (3D). In one example embodiment, the method comprises capturing a series of successive 3D images in real time, identifying that the object has a predetermined elongated shape, identifying that the object is oriented substantially towards a predetermined direction, determining at least one qualifying action being performed by a user and/or the object, comparing the at least one qualifying action to one or more pre-determined actions associated with the direction towards which the object is oriented, and, based on the comparison, selectively issuing to the one or more electronic devices a command associated with the at least one qualifying action.

Abstract: Provided are computer-implemented methods and systems for controlling devices using gestures and a 3D sensor that enables implementing the above. In one embodiment, the method proposed herein may be based on defining at least one sensor area within the space surrounding a user of a controlled device; associating this sensor area with at least one user gesture; associating the combination of the user gesture and sensor area with an actionable command; identifying the direction of the line of sight of the user and a focal point of the line of sight of the user; and, if the line of sight of the user is directed a sensor area, issuing an actionable command corresponding to the combination of the sensor area and the gesture that the user makes while looking at this sensor area.

Abstract: A computer-implemented method and system for controlling various electronic devices by recognition of gestures made by a user within a particular space defined in front of the user are provided. An example method may comprise generating a depth map of a physical scene, determining that a head of the user is directed towards a predetermined direction, establishing a virtual sensing zone defined between the user and a predetermined location, identifying a particular gesture made by the user within the virtual sensing zone, and selectively providing to the electronic device a control command associated with the particular gesture. The particular gesture may be performed by one or more characteristic forms provided by the user within the virtual sensing zone being in an active state. The characteristic forms are forms reliably distinguishable from casual forms by means of computer vision and having certain attributes, which can reliably reflect user intent.