Reconstruction of incomplete decoded videos for use in objective quality metrics

Many full-reference objective metrics require that the original and the degraded video contain the same number of frames. Most codecs are not able to decode the video properly when the video is subject to packet losses and produce incomplete video files with lower number of frames than the original video. In this paper we present a simple method to reconstruct the degraded videos so that it has the same length as the original. Information regarding the frame numbers is inserted into the original uncompressed video and used later on to identify the missing frames and reconstruct the degraded video.