The present disclosure relates to a method and device for using latency compensatory pose prediction with respect to three-dimensional (3D) media data to perform rendering in a communication system supporting mixed reality (XR)/augmented reality (AR). According to an embodiment of the present disclosure, a method for a first device, which receives 3D media data from a media server in a communication system, to perform rendering comprises the steps of: receiving, from AR glasses, pose prediction-related information including pose information of a first point in time; performing pose prediction of a second point in time, at which 2-dimensional (2D) rendering is to be performed in the AR glasses, on the basis of the pose prediction-related information; rendering one or a plurality of 2D pose prediction rendering views with respect to the received 3D media data on the basis of one or a plurality of pieces of predicted pose information of the second point in time; and transmitting, to the AR glasses, 2D media data compressed by encoding the one or plurality of 2D pose prediction rendering views.