FREDSR: Fourier Residual Efficient Diffusive GAN for Single Image Super Resolution
This work addresses efficient super-resolution for real-time applications, but it is incremental as it builds on existing GAN methods with dataset-specific optimizations.
The paper tackled single image super-resolution by proposing FREDSR, a GAN variant that achieved strong performance on the UHDSR4K dataset for 3x upscaling from 360p and 720p with only 37,000 parameters, though it sacrifices generalizability.
FREDSR is a GAN variant that aims to outperform traditional GAN models in specific tasks such as Single Image Super Resolution with extreme parameter efficiency at the cost of per-dataset generalizeability. FREDSR integrates fast Fourier transformation, residual prediction, diffusive discriminators, etc to achieve strong performance in comparisons to other models on the UHDSR4K dataset for Single Image 3x Super Resolution from 360p and 720p with only 37000 parameters. The model follows the characteristics of the given dataset, resulting in lower generalizeability but higher performance on tasks such as real time up-scaling.