Official code for ECCV2022 paper: Learning Series-Parallel Lookup Tables for Efficient Image Super-Resolution
The folder training_testing_code contains the code for training SPLUT, transferring mapping modules to LUTs, and testing SPLUT models in the training_testing_code folder.
The users can follow the descriptions in readme.md to train and test the SP-LUT models.
In order to evaluate the SR efficiency on mobile phones, we have designed demos for SPLUT and SR-LUT. The code is also released and is in the demo code folder. By following the guides in readme_demo.md, the users can generate the corresponding APK files for testing.