A method and device for synthesizing an image for audiovisual communication includes: generating, if a request for synthesizing a sensor image input through a camera sensor and an image designated by a user is input, a synthesis image of the sensor image and the designated image. The method also includes storing, when the synthesis image is generated, object information of an object of the sensor image and object information of an object of the designated image; and changing, if a sensor image is input in which object information of the sensor image object has changed, object information of the designated image object according to the changed object information of the sensor image object and synthesizing the sensor image and the designated image in which object information has changed.