Corpus ID: 220363410

Can Un-trained Neural Networks Compete with Trained Neural Networks at Image Reconstruction?

  title={Can Un-trained Neural Networks Compete with Trained Neural Networks at Image Reconstruction?},
  author={Mohammad Zalbagi Darestani and Reinhard Heckel},
Convolutional Neural Networks (CNNs) are highly effective for image reconstruction problems. Typically, CNNs are trained on large amounts of training images. Recently, however, un-trained neural networks such as the Deep Image Prior and Deep Decoder have achieved excellent image reconstruction performance for standard image reconstruction problems such as image denoising and image inpainting, without using any training data. This success raises the question whether un-trained neural networks… Expand
  • 2020
Deep neural networks have emerged as very successful tools for image restoration and reconstruction tasks. These networks are often trained end-to-end to directly reconstruct an image from a noisy orExpand
Data augmentation for deep learning based accelerated MRI reconstruction with limited data
The DA pipeline, MRAugment, is specifically designed to utilize the invariances present in medical imaging measurements as naive DA strategies that neglect the physics of the problem fail, and demonstrates that in the low-data regime DA prevents overfitting and can match or even surpass the state of the art while using significantly fewer training data. Expand


Compressive sensing with un-trained neural networks: Gradient descent finds the smoothest approximation
It is shown that---without any further regularization---an un-trained convolutional neural network can approximately reconstruct signals and images that are sufficiently structured, from a near minimal number of random measurements. Expand
Deep Decoder: Concise Image Representations from Untrained Non-convolutional Networks
This paper proposes an untrained simple image model, called the deep decoder, which is a deep neural network that can generate natural images from very few weight parameters, with a simple architecture with no convolutions and fewer weight parameters than the output dimensionality. Expand
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, superresolution, and inpainting. Expand
A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction
This work proposes a framework for reconstructing MR images from undersampled data using a deep cascade of convolutional neural networks to accelerate the data acquisition process and shows that for Cartesian undersampling of 2D cardiac MR images, the proposed method outperforms the state-of-the-art compressed sensing approaches. Expand
Denoising and Regularization via Exploiting the Structural Bias of Convolutional Generators
A step towards demystifying this experimental phenomenon is taken by attributing this effect to particular architectural choices of convolutional networks, namely convolutions with fixed interpolating filters, and it is proved that early-stopped gradient descent denoises/regularizes. Expand
U-Net: Convolutional Networks for Biomedical Image Segmentation
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Expand
Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors
It is shown both theoretically as well as empirically that with deep network priors, one can achieve better compression rates for the same image quality compared to hand crafted priors. Expand
Pyramid Convolutional RNN for MRI Reconstruction
A novel deep learning-based method, Pyramid Convolutional RNN (PC-RNN), to reconstruct the image from multiple scales, and the results show that the proposed model achieves significant improvements than other methods and can recover more fine details. Expand
GrappaNet: Combining Parallel Imaging With Deep Learning for Multi-Coil MRI Reconstruction
The proposed method, called GrappaNet, performs progressive reconstruction by first mapping the reconstruction problem to a simpler one that can be solved by a traditional parallel imaging methods using a neural network, followed by an application of a parallel imaging method, and finally fine-tuning the output with another neural network. Expand
Image denoising: Can plain neural networks compete with BM3D?
This work attempts to learn this mapping directly with a plain multi layer perceptron (MLP) applied to image patches and shows that by training on large image databases it is able to compete with the current state-of-the-art image denoising methods. Expand