ROBUST SELF-GUIDED DEEP IMAGE PRIOR

In this work, we study the deep image prior (DIP) for reconstruc- tion problems in magnetic resonance imaging (MRI). DIP has be- come a popular approach for image reconstruction, where it recovers the clear image by fitting an overparameterized convolutional neu- ral network (CNN) to the corrupted/undersampled measurements. To improve the performance of DIP, recent work shows that using a reference image as an input often leads to improved reconstruc- tion results compared to vanilla DIP with random input. However, obtaining the reference input image often requires supervision and hence is difficult in practice. In this work, we propose a self-guided reconstruction scheme that uses no training data other than the set of undersampled measurements to simultaneously estimate the network weights and input (reference). We introduce a new regularization that aids the joint estimation by requiring the CNN to act as a pow- erful denoiser. The proposed self-guided method gives significantly improved image reconstructions for MRI with limited measurements compared to the conventional DIP and the reference-guided method while eliminating the need for any additional data.