Apparatus and method for configuring a screen for a video call using a facial expression by recognizing a face from an image, calculating facial expression information for an expression of the recognized face, and determining whether there is a change in expression of the recognized face by comparing the calculated facial expression information with reference expression information preset to determine a change in expression of the face. If there is a change in expression of the recognized face, the apparatus and method selects a video image corresponding to the changed expression in the video call screen, and reconfigures the video call screen using the selected video image, making it possible for a user to conveniently select an image of the interested person without taking extensive action, and preventing a wrong image from being selected due to the unintended user facial movement.