Abstract: A gesture control system is provided having multiple attention levels. The gesture control system monitors for events based on the current attention level that it is in, while being free to ignore events at other attention levels. In an initial attention level, the gesture control system may monitor for an event to cause it to transition to an active state comprising a second attention level. In the second attention level, the gesture control system may monitor for a user gesture to perform an action on an electronic device. Upon detecting the user gesture, the gesture control system may transition to a third attention level where it monitors for a voice command or other input that modifies the meaning of the user gesture. The gesture control system may then perform an action based on the user gesture and the voice command or other input.

Abstract: A method and system may be used to determine gestures of one or more users from a video. Motion may be detected in an image frame of a video, and the image frame may be cropped around the motion. Body pose estimation may be performed on the cropped image frame. The location of the user's hands may be determined from the body pose. Additional processing may be performed to identify hand gestures.

Abstract: A method and system may be used to provide a user interface allowing a user to identify electronic devices by placing bounding boxes on a video of scene. The bounding boxes may identify one or more electronic devices. Coordinates of the bounding boxes may be stored to allow determining when a user is gesturing towards one of the bounding boxes. The coordinates of the bounding boxes may be updated by a computer vision process that determines whether one or more electronic devices have been moved. When it is detected that a user is gesturing towards a bounding box, an electronic device associated with the bounding box may be controlled.