2021/11/21 by Maneesh John, John, Maneesh, Hemant Kumar Aggarwal +5
Earth and Planetary Sciences · Computer Science · Medicine · #Seismic Imaging and Inversion Techniques #Advanced Image Processing Techniques #Medical Imaging Techniques and Applications
paper · pdf · doi:10.48550/arxiv.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. We introduce a generalized Stein's unbiased risk estimate (GSURE) loss metric to minimize the overfitting. Our experiments show that the SURE-DIP approach minimizes the overfitting issues, thus offering significantly improved performance over classical DIP schemes. We also use the SURE-DIP approach with model-based unrolling architectures, which offers improved performance over direct inversion schemes.