Arithmetic encoding is an essential class of coding techniques which have been widely used in various data compression systems and exhibited promising performance. One key issue of arithmetic encoding method is to predict the probability of the current symbol to be encoded from its context, i.e., the preceding encoded symbols, which usually can be executed by building a look-up table (LUT). However, the complexity of LUT increases exponentially with the length of context. Thus, such solutions are limited in modeling large context, which inevitably restricts the compression performance. Several recent convolutional neural network (CNN) and recurrent neural network (RNN)-based solutions have been developed to account for large context, but are still costly in computation. The inefficiency of the existing methods are mainly attributed to that probability prediction is performed independently for the neighboring symbols, which actually can be efficiently conducted by shared computation. To this end, we propose a trimmed convolutional network for arithmetic encoding (TCAE) to model large context while maintaining computational efficiency. As for trimmed convolution, the convolutional kernels are specially trimmed to respect the compression order and context dependency of the input symbols. Benefited from trimmed convolution, the probability prediction of all symbols can be efficiently performed in one single forward pass via a fully convolutional network. Experiments show that our TCAE attains better compression ratio in lossless gray image compression, and can be adopted in CNN-based lossy image compression to achieve state-of-the-art rate-distortion performance with real-time encoding speed.
View on arXiv