DIPNMF: Nonnegative Matrix Factorization with Deep Image Prior [ICCV2019]
This is an implementation of below ICCV2019 paper in Python with TensorFlow.
"Dynamic PET Image Reconstruction Using Nonnegative Matrix Factorization Incorporated With Deep Image Prior"
Tatsuya Yokota, Kazuya Kawai, Muneyuki Sakata, Yuichi Kimura, Hidekata Hontani.
International Conference on Computer Vision (ICCV), 2019.
First, please download a file ("https://drive.google.com/open?id=1khMgXQ-c3imWFWUx2n3vuvb56jE_kCWx"), and put this file on the same folda of the code. This file includes, simulation dynamic PET, noisy sinograms, and a Radon transform matrix.
This code was written by Tatsuya Yokota.