An Educated Warm Start for Deep Image Prior-Based Micro CT Reconstruction

Deep image prior (DIP) was recently introduced as an effective unsupervised approach for restoration tasks. DIP represents the to be recovered output of a deep convolutional neural network, and learns network's parameters such that matches corrupted observation. Despite its impressive reconstructive properties, is slow when compared supervisedly learned, or traditional reconstruction techniques. To address computational challenge, we bestow with two-stage learning paradigm: (i) perform supervised pretraining network on simulated dataset; (ii) fine-tune adapt target task. We provide thorough empirical analysis shed insights into impacts in context reconstruction. showcase considerably speeds up stabilizes subsequent task from real-measured 2D 3D micro computed tomography data biological specimens. The code additional experimental materials are available at https://educateddip.github.io/docs.educated_deep_image_prior/.