TY - GEN
T1 - Natural image denoising with convolutional networks
AU - Jain, Viren
AU - Seung, H. Sebastian
PY - 2009
Y1 - 2009
N2 - We present an approach to low-level vision that combines two main ideas: the use of convolutional networks as an image processing architecture and an unsupervised learning procedure that synthesizes training samples from specific noise models. We demonstrate this approach on the challenging problem of natural image denoising. Using a test set with a hundred natural images, we find that convolutional networks provide comparable and in some cases superior performance to state of the art wavelet and Markov random field (MRF) methods. Moreover, we find that a convolutional network offers similar performance in the blind denoising setting as compared to other techniques in the non-blind setting. We also show how convolutional networks are mathematically related to MRF approaches by presenting a mean field theory for an MRF specially designed for image denoising. Although these approaches are related, convolutional networks avoid computational difficulties in MRF approaches that arise from probabilistic learning and inference. This makes it possible to learn image processing architectures that have a high degree of representational power (we train models with over 15,000 parameters), but whose computational expense is significantly less than that associated with inference in MRF approaches with even hundreds of parameters.
AB - We present an approach to low-level vision that combines two main ideas: the use of convolutional networks as an image processing architecture and an unsupervised learning procedure that synthesizes training samples from specific noise models. We demonstrate this approach on the challenging problem of natural image denoising. Using a test set with a hundred natural images, we find that convolutional networks provide comparable and in some cases superior performance to state of the art wavelet and Markov random field (MRF) methods. Moreover, we find that a convolutional network offers similar performance in the blind denoising setting as compared to other techniques in the non-blind setting. We also show how convolutional networks are mathematically related to MRF approaches by presenting a mean field theory for an MRF specially designed for image denoising. Although these approaches are related, convolutional networks avoid computational difficulties in MRF approaches that arise from probabilistic learning and inference. This makes it possible to learn image processing architectures that have a high degree of representational power (we train models with over 15,000 parameters), but whose computational expense is significantly less than that associated with inference in MRF approaches with even hundreds of parameters.
UR - http://www.scopus.com/inward/record.url?scp=78149296699&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78149296699&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:78149296699
SN - 9781605609492
T3 - Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference
SP - 769
EP - 776
BT - Advances in Neural Information Processing Systems 21 - Proceedings of the 2008 Conference
PB - Neural Information Processing Systems
T2 - 22nd Annual Conference on Neural Information Processing Systems, NIPS 2008
Y2 - 8 December 2008 through 11 December 2008
ER -