A display unit 11 has a display area of a predetermined shape and displays an image on the display area. An image capturing unit 12 generates data of a captured image by capturing an image of an eye 2 of a user in which the display area is reflected. A reference detection unit 21 detects a moving reference point that moves along with the movement of user's eye-gaze and an unmoving reference point that can be assumed as approximately stationary regardless of the user's eye-gaze movement, and generates a vector drawn from the unmoving reference point to the moving reference point as a shift amount V(x,y). An eye-gaze detection unit 22 detects a movement vector ΔV(Δx,Δy) as the user's eye-gaze movement amount based on a reference shift amount Vo(xo,yo) generated in the past and the shift amount V(x,y) currently generated.