Depth camera-based relative gesture detection

摘要

The subject disclosure is directed towards using timing and/or relative depth data to reduce false positives in gesture detection within a depth-sensed region. Depth camera data is processed to position a cursor over a displayed representation of a control. If the user's hand hovers over the control for a threshold time period, and then changes hand depth a relative amount (e.g., pushes the hand forward a delta amount), an event is fired. The displayed representation of the control may change (e.g., enlarge) upon hovering. The relative depth may be computed based upon the depth when the user hand initially enters the control area. The relative depth may remain the same if the user pulls the hand away from the camera by tracking the maximum depth and firing the event when the maximum depth value minus the current depth value reaches a delta value.

主权项

1. In a computing environment, a method performed at least in part on at least one processor, comprising:
receiving depth data associated with one or more frames of pixels, a pixel having a depth coordinate; processing the depth data with respect to a depth-sensed region to determine a first depth coordinate based upon an object being present in a sub-region that corresponds to a displayed representation of a control; changing at least one visible property of the displayed representation of the control based upon a hover time of the object in the sub-region corresponding to a location of the displayed representation of the control; determining whether one or more criteria are met, the one or more criteria including the hover time in which the object remains present in the sub-region and further depth data that indicates whether the object has moved to a second depth coordinate that is relative to the first depth coordinate, said determining comprising tracking a maximum depth of the object by the control, wherein the control tracks the depth data for any gesture made relative to the control, the second depth coordinate is a current depth and the first depth coordinate is the maximum depth, and wherein a delta change from the maximum depth to the current depth fires an event, and wherein the control comprises a plurality of keys, and further comprising: tracking the maximum depth for any gesture made relative to a key in the plurality of keys; and responsive to the determination that the one or more criteria are met, firing the event.