pytorch based Artifact Reduction CNN. This CNN is based on Chao Dong, et al. “Compression Artifacts Reduction by a Deep Convolutional Network.” ICCV 2015 and Pavel Svoboda and et al. “Compression Artifacts Removal Using Convolutional Neural Networks.” ICCV 2017
IVC_lab final optimization project in 18/19SS, Technische Universitaet Muenchen.
open ARCNN.ipynb with Jupyter notebook, and follow the instructions.
the classes including CNN model and dataset are in the "Functions_and_Clsses.py"
in this folder we only keep one test sequences: "foreman20_40_RGB"