by Mario González, Javier Preciozzi, Pablo Musé, Andrés Almansa
Abstract:
Wavelet compression schemes such as JPEG2000 may lead to very specific visual artifacts due to quantization of noisy wavelet coefficients. These artifacts have highly spatially-correlated structure, making it difficult to be removed with standard denoising algorithms. In this work, we propose a joint denoising and decompression method that combines a data-fitting term, which takes into account the quantization process, and an implicit prior learnt using a state-of-the-art denoising CNN.
Reference:
Joint denoising and decompression using CNN regularization (Mario González, Javier Preciozzi, Pablo Musé, Andrés Almansa), In CVPR Workshop and Challenge on Learned Image Compression, 2018.
Bibtex Entry:
@inproceedings{Gonzalez2018-CVPR,
Abstract = {Wavelet compression schemes such as JPEG2000 may lead to very specific visual artifacts due to quantization of noisy wavelet coefficients. These artifacts have highly spatially-correlated structure, making it difficult to be removed with standard denoising algorithms. In this work, we propose a joint denoising and decompression method that combines a data-fitting term, which takes into account the quantization process, and an implicit prior learnt using a state-of-the-art denoising CNN.},
Address = {Salt Lake City, Utah, United States.},
Author = {Gonz{\'{a}}lez, Mario and Preciozzi, Javier and Mus{\'{e}}, Pablo and Almansa, Andr{\'{e}}s},
Booktitle = {CVPR Workshop and Challenge on Learned Image Compression},
Month = {jun},
Pages = {2598--2601},
Title = {{Joint denoising and decompression using CNN regularization}},
Url = {https://hal.archives-ouvertes.fr/hal-01825573},
Year = {2018},
Bdsk-Url-1 = {https://hal.archives-ouvertes.fr/hal-01825573}}