Performance optimized predictor blending technique for lossless image coding

The paper presents a lossless coding method based on predictor blending approach that codes nine widely used benchmark images with the lowest average bitrate ever published. At the same time the algorithm is more time efficient than its most important competitors: TMW
^LEGO
, MRP 0.5, and Multi-WLS. A set of 20 blended predictors is applied, including adaptive RLS/OLS ones, ALCM, CoBALP, and a texture context mapping one. A sophisticated error bias removal method is applied to outputs from few of them. Prediction error is coded using an advanced adaptive context arithmetic coder.