Chang Sun

TL;DR: This work proposes an efficient statistical-model-based algorithm using the Laplacian model to design quantization tables for JPEG encoding and shows that a quantization table can be optimized in a way that the resulting distortion complies with certain behavior.

TL;DR: This work combines dithering with soft decision quantization (SDQ)—yielding dithered SDQ, and an iterative algorithm is proposed for jointly designing dithers for DCT coefficients at each frequency, achieving comparable and sometimes better rate-distortion performance with 65% computational complexity reduction.

TL;DR: This paper proposes an efficient DCT-based non-predictive image compression system, where quantization and entropy coding are completely re-designed based on the Laplacian transparent composite model, which achieves the best rate distortion (RD) coding performance among all non-Predictiveimage compression systems proposed in the literature.

TL;DR: An improved iterative algorithm is first proposed to calculate the rate-distortion performance of causal video coding for any continuous sources and iteratively updates them along with transitional probabilities from the continuous source to reproduction letters, thus overcoming the computation complexity problem encountered when applying the algorithm recently proposed by Yang et al for discrete sources to continuous sources.