• Corpus ID: 244478075

Deep Image Prior using Stein's Unbiased Risk Estimator: SURE-DIP

@article{John2021DeepIP,
  title={Deep Image Prior using Stein's Unbiased Risk Estimator: SURE-DIP},
  author={Maneesh John and Hemant Kumar Aggarwal and Qing Zou and Mathews Jacob},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.10892}
}
Deep learning algorithms that rely on extensive training data are revolutionizing image recovery from ill-posed measurements. Training data is scarce in many imaging applications, including ultra-high-resolution imaging. The deep image prior (DIP) algorithm was introduced for single-shot image recovery, completely eliminating the need for training data. A challenge with this scheme is the need for early stopping to minimize the overfitting of the CNN parameters to the noise in the measurements… 
1 Citations

Figures from this paper

Pyramid Convolutional RNN for MRI Image Reconstruction

A novel deep learning based method, Pyramid Convolutional RNN (PC-RNN), to reconstruct images from multiple scales, based on the formulation of MRI reconstruction as an inverse problem, with results showing that the proposed model outperforms other methods and can recover more details.

References

SHOWING 1-10 OF 10 REFERENCES

Ensure: Ensemble Stein’s Unbiased Risk Estimator for Unsupervised Learning

  • H. AggarwalM. Jacob
  • Computer Science
    ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2021
It is shown that training a network using an ensemble of images, each acquired with a different sampling pattern, can closely approximate the mean square error (MSE) estimate of Stein’s Unbiased Risk Estimator.

Training Deep Learning Based Image Denoisers From Undersampled Measurements Without Ground Truth and Without Image Prior

This work proposes novel methods based on two well-grounded theories: denoiser-approximate message passing (D-AMP) and Stein's unbiased risk estimator (SURE), able to train deep learning based image denoisers from undersampled measurements without ground truth images and without additional image priors, and to recover images with state-of-the-art qualities fromundersampled data.

Unsupervised Learning with Stein's Unbiased Risk Estimator

It is shown that Stein's Unbiased Risk Estimator (SURE) and its generalizations can be used to train convolutional neural networks (CNNs) for a range of image denoising and recovery problems without any ground truth data.

MoDL: Model-Based Deep Learning Architecture for Inverse Problems

This work introduces a model-based image reconstruction framework with a convolution neural network (CNN)-based regularization prior, and proposes to enforce data-consistency by using numerical optimization blocks, such as conjugate gradients algorithm within the network.

Deep Image Prior

It is shown that a randomly-initialized neural network can be used as a handcrafted prior with excellent results in standard inverse problems such as denoising, super-resolution, and inpainting.

A Bayesian Perspective on the Deep Image Prior

It is shown that the deep image prior is asymptotically equivalent to a stationary Gaussian process prior in the limit as the number of channels in each layer of the network goes to infinity, and derive the corresponding kernel, which informs a Bayesian approach to inference.

Dynamic Imaging Using a Deep Generative SToRM (Gen-SToRM) Model

A generative smoothness regularization on manifolds (SToRM) model for the recovery of dynamic image data from highly undersampled measurements and introduces an efficient progressive training-in-time approach and an approximate cost function.

Generalized SURE for Exponential Families: Applications to Regularization

A regularized SURE objective is proposed, and its use in the context of wavelet denoising is demonstrated, and a new method for choosing regularization parameters in penalized LS estimators is suggested.

Monte-Carlo Sure: A Black-Box Optimization of Regularization Parameters for General Denoising Algorithms

A novel Monte-Carlo technique is presented which enables the user to calculate SURE for an arbitrary denoising algorithm characterized by some specific parameter setting and it is demonstrated numerically that SURE computed using the new approach accurately predicts the true MSE for all the considered algorithms.

Estimation of the Mean of a Multivariate Normal Distribution