We have proposed in this project a new image model that combines the non-local means and sparse coding approaches to image restoration into a unified framework where similar patches are decomposed using similar sparsity patterns.
Quantitative and qualitative experiments with images corrupted with synthetic or real noise have shown that the proposed algorithm outperforms the state of the art in image demosaicking and denoising tasks. Next on our agenda is to include non-uniform noise models in the reconstruction process, then adapt our approach to other challenging image manipulation problems in computational photography, including deblurring, inpainting, and texture synthesis in still images and video sequences.
Mairal, Julien, Francis Bach, Jean Ponce, Guillermo Sapiro, and Andrew Zisserman. “Non-local sparse models for image restoration.” In Computer Vision, 2009 IEEE 12th International Conference on, pp. 2272-2279. IEEE, 2009.