Image compression using the Iteration-Tuned and Aligned Dictionary

We present a new, block-based image codec based on sparse representations using a learned, structured dictionary called the IterationTuned and Aligned Dictionary (ITAD). The question of selecting the number of atoms used in the representation of each image block is addressed with a new, global (image-wide), rate-distortion-based sparsity selection criterion. We show experimentally that our codec outperforms JPEG2000 in both quantitative evaluations (by 0.9 dB to 4 dB) and qualitative evaluations.

• Type: Journal
• Year: 2011
• Where: ICASSP
• Authors: Joaquin Zepeda, Christine Guillemot, Ewa Kijak
• Keywords: Block-based Image Codec, ICASSP 2011, Signal Processing, Sparse Representations, Sparsity Selection Criterion