A multi-view image display apparatus is provided. The apparatus includes a crosstalk compensator configured to compensate for crosstalk with respect to multi-view images by using a crosstalk matrix defining the crosstalk that occurs between the multi-view images; a tracker configured to track a position of a user's face as a user's position is moved; a controller configured to determine at least two views to be provided as images for user's left and right eyes from among the multi-view images based on the tracked position of the user's face, and to control the crosstalk compensator to compensate for the crosstalk by applying a predetermined weight to the at least two determined views; and a display configured to display the multi-view images of which the crosstalk is compensated.