An inputter inputs a captured image of a hand captured by a camera. An acquirer acquires information indicating the distance between the camera and the hand. A storer stores reference data for specifying a hand gesture and a command corresponding to the gesture, for each distance between the camera and the hand. A selector selects, from among the reference data stored in the storer, reference data corresponding to the distance indicated by the information acquired by the acquirer. A specifier refers to the reference data selected by the selector, specifies a hand gesture in the captured image input by the inputter, and specifies a command corresponding to the specified gesture.