TY - GEN
T1 - High Perceptual Quality Image Denoising with a Posterior Sampling CGAN
AU - Ohayon, Guy
AU - Adrai, Theo
AU - Vaksman, Gregory
AU - Elad, Michael
AU - Milanfar, Peyman
N1 - Publisher Copyright: © 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - The vast work in Deep Learning (DL) has led to a leap in image denoising research. Most DL solutions for this task have chosen to put their efforts on the denoiser's architecture while maximizing distortion performance. However, distortion driven solutions lead to blurry results with sub-optimal perceptual quality, especially in immoderate noise levels. In this paper we propose a different perspective, aiming to produce sharp and visually pleasing denoised images that are still faithful to their clean sources. Formally, our goal is to achieve high perceptual quality with acceptable distortion. This is attained by a stochastic denoiser that samples from the posterior distribution, trained as a generator in the framework of conditional generative adversarial networks (CGAN). Contrary to distortion-based regularization terms that conflict with perceptual quality, we introduce to the CGAN objective a theoretically founded penalty term that does not force a distortion requirement on individual samples, but rather on their mean. We showcase our proposed method with a novel denoiser architecture that achieves the reformed denoising goal and produces vivid and diverse outcomes in immoderate noise levels.
AB - The vast work in Deep Learning (DL) has led to a leap in image denoising research. Most DL solutions for this task have chosen to put their efforts on the denoiser's architecture while maximizing distortion performance. However, distortion driven solutions lead to blurry results with sub-optimal perceptual quality, especially in immoderate noise levels. In this paper we propose a different perspective, aiming to produce sharp and visually pleasing denoised images that are still faithful to their clean sources. Formally, our goal is to achieve high perceptual quality with acceptable distortion. This is attained by a stochastic denoiser that samples from the posterior distribution, trained as a generator in the framework of conditional generative adversarial networks (CGAN). Contrary to distortion-based regularization terms that conflict with perceptual quality, we introduce to the CGAN objective a theoretically founded penalty term that does not force a distortion requirement on individual samples, but rather on their mean. We showcase our proposed method with a novel denoiser architecture that achieves the reformed denoising goal and produces vivid and diverse outcomes in immoderate noise levels.
UR - http://www.scopus.com/inward/record.url?scp=85123049393&partnerID=8YFLogxK
U2 - https://doi.org/10.1109/ICCVW54120.2021.00207
DO - https://doi.org/10.1109/ICCVW54120.2021.00207
M3 - منشور من مؤتمر
T3 - Proceedings of the IEEE International Conference on Computer Vision
SP - 1805
EP - 1813
BT - Proceedings - 2021 IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021
T2 - 18th IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021
Y2 - 11 October 2021 through 17 October 2021
ER -