Image compression using the Iteration-Tuned and Aligned Dictionary

Joaquin Zepeda (1, 2) , Christine Guillemot (1) , Ewa Kijak (2)

1 TEMICS - Digital image processing, modeling and communication
2 TEXMEX - Multimedia content-based indexing

We present a new, block-based image codec based on sparse representations using a learned, structured dictionary called the Iteration-Tuned 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-distorition-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.

Keywords: Image compression; sparse representations; learned dictionaries; structured dictionaries; matching pursuit

Domains: Image Processing [eess.IV]; Signal and Image Processing; Signal and Image processing

Joaquin Zepeda, Christine Guillemot, Ewa Kijak. Image compression using the Iteration-Tuned and Aligned Dictionary. IEEE International Conference on Acoustics, Speech and Signal Processing, May 2011, Czech Republic. pp.793 - 796, ⟨10.1109/ICASSP.2011.5946523⟩. ⟨hal-00647253⟩