A location of a first portion of a hand and a location of a second portion of the hand are detected within a working volume, the first portion and the second portion being in a horizontal plane. A visual representation is positioned on a display based on the location of the first portion and the second portion. A selection input is initiated when a distance between the first portion and the second portion meets a predetermined threshold, to select an object presented on the display, the object being associated with the location of the visual representation. A movement of the first portion of the hand and the second portion of the hand also may be detected in the working volume while the distance between the first portion and the second portion remains below the predetermined threshold and, in response, the object on the display can be repositioned.