Super-resolution using a CNN, based on the work of the DGtal team. First of all, an Nvidia graphics card (neither AMD nor Intel integrated) is highly recommended to parallelize the CNN. You will then need to install CUDA. No CUDA = dozens of times slower. This program will generate "model_epoch_ .pth" files corresponding to the model at epoch n, in a folder saved_model_u t_bs bs_tbs tbs_lr lr, where corresponds to the scale factor, bsthe size of the training batch, tbsthe size of the test batch and lrto the learning rate. Low res images should be located in a "dataset/input" folder, and high res targets in a "dataset/target" folder, where each different quality image has the same name in both folders.
Features
- GTVImageVect (C++ to compile or Linux binary to download)
- ImageMagick (Available via sudo apt install imagemagick)
- Generally speaking, a Linux, emulator or WSL
- An Nvidia graphics card (neither AMD nor Intel integrated) is highly recommended
- Parallelize the CNN
- Based on the work of the DGtal team
Project Samples
Categories
Video UpscalersLicense
GNU General Public License version 3.0 (GPLv3)Follow Super-résolution via CNN
You Might Also Like
Relax: PRTG Monitors Your IT for You
You’re the go-to IT person, always putting out fires and keeping things running. With PRTG, you get reliable alerts to monitor your entire IT infrastructure, without the noise. Our intuitive setup gives you a clear overview of your network, devices, and applications in real time. Get instant alerts only when something needs your attention, whether you’re at your desk or on the move. Spend less time worrying about outages and more time focusing on what matters. Set up PRTG once and let it work for you - PRTG has you covered.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of Super-résolution via CNN!