Exploitation of context classification for parallel pixel coding in JPEG-LS

Many state-of-the-art lossless image compression standards feature adaptive error modelling. This, however, leads to data dependency loops of the compression scheme such that a parallel compression of neighboring pixels is not possible. In this paper, we analyse the parallelism provided in JPEG-LS by the large number of different and independent context sets. An out-of-order scheduling algorithm is introduced selecting independent pixels classified to be in different contexts. These independent pixels are processed in parallel. The compressed data is fully compatible with the ISO/IEC 14495-1 JPEG-LS standard. A prototype implementation on a modern GPU is provided to verify an overall speedup.