Provided are an apparatus and method for controlling a virtual mouse based on a hand motion. The apparatus includes an image processing unit configured to measure a maximum horizontal width of a hand region of a user and a width of an index finger in any one of a first depth image captured in a state in which the index finger is spread out in a vertical direction with his or her fist and a second depth image captured in a state in which the index finger is folded with the fist, a hand motion recognizing unit configured to compare the maximum horizontal width and the width of the index finger to recognize whether the index finger is folded, and a function matching unit configured to match a change in a folded state of the index finger to a pre-set first function and output a control signal for the function.