taqu / cppraisr

RAISR: superresolution C++ implementation

cppraisr

A C++ implementation of RAISR.

Prerequisites

• CMake 3.18<=

Building

Windows

$ mkdir build & cd build
$ cmake .. -G"Visual Studio 17 2022"

Linux

$ mkdir build && cd build
$ cmake .. && make

Training

Put images for traing in the train_data directory.

$ ./bin/train

You can use some options.

usage: train [-help] [-f Filter] [-max Max Images] [-threads Number of Threads]
arguments:
        -help   show this help
        -f      a filter name to write without extension
        -max    the max number of training images
        -threads        the number of threads to use

Testing

./bin/test tests images under the test_data.

usage: test [-help] [-f Filter] [-max Max Images] [-q]
arguments:
        -help   show this help
        -f      a filter to use
        -max    the max number of testing images
        -q      switch to quality testing mode, measure the avarage of each MSSIMs.

References

• Y. Romano, J. Isidoro and P. Milanfar, "RAISR: Rapid and Accurate Image Super Resolution" in IEEE Transactions on Computational Imaging, vol. 3, no. 1, pp. 110-125, March 2017.

License

This software is distributed under two licenses, MIT License or Public Domain, choose whichever you like.