An empirical study to turn MPEG-2 into lossless video compression

While digital video compression is predominantly lossy, we report a basic algorithm design for lossless video compression from schemes designed in MPEG-2, JPEG-LS and arithmetic coding. The designed algorithm features in (a) macro-block based motion estimation and compensation; (b) intra-frame coding via JPEG-LS; and (c) arithmetic coding. As there is little work reported in the area of lossless video compression, we assess the designed algorithm in comparison with the state-of-the-art lossless image compression including JPEG-LS and CALIC. Extensive experiments support that one of the designed algorithms outperform all the benchmarks and thus providing a good platform for further research and development.