Transcoding of MPEG-2 video

Abstract

Transcoding is a technique which can compress a video either without loss of quality or with loss of quality. Transcoding with loss of quality can obviously provide higher compression rates. In some way or another, everyone has experienced firsthand the importance of the size of data. Naturally, the smaller the better, since one would be able to store more data on one's preferred memory device. Several standards exist for coding audio-visual information into a digital compressed format. These include the popular MPEG-2. The development of new coding standards aims to fulfil the shortcomings of preceding standards, and provide more quality while availing less memory. The MPEG-2 coding method is wide spread and very commonly used. The only problem when using this compression standard is the space required to store the video. Hence, the main motivator for this project is to answer the research question 'can MPEG-2 encoded video be further compressed, without losing any data and maintaining the same quality?' This thesis is about transcoding of MPEG-2 video. Several solutions were studied in order to understand how changes can be applied successfully for compression, all the while taking into consideration the video quality since it constitutes an important part. Such studies incorporate typical statistics of MPEG-2 video files. Several tests were carried out and noted. Negative effects were also documented, and research was done to check the feasibility. MPEG-2 video was compressed by removing redundant frames. This is in itself a lossy type of compression, but for each step of frame dropping done, care was taken such that the quality of the video remained unaltered, or if there was any loss, it was inapprehensible.