Adaptive Prediction, Context Modeling, and Entropy Coding Methods for CALIC Lossless Image Compression

The three main pillars for lossless image compression are prediction, prediction residual correction and entropy coding. In this work, we propose a lossless image compression algorithm which utilizes local information to dynamically switch between different prediction techniques. Moreover, we improve on traditional prediction techniques and modify them into diagonal GAP and dynamic window weighted linear prediction. Finally, since the coding efficiency is directly proportional to the context being chosen if context arithmetic coding is applied, we propose a robust context model which utilizes local information to get better coding efficiency.