A wavelet diffusion GAN for SISR reduces diffusion timesteps via the diffusion GAN paradigm and applies DWT for dimensionality reduction, claiming faster training/inference and higher fidelity than prior methods on CelebA-HQ.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Wavelet Diffusion GAN for Image Super-Resolution
A wavelet diffusion GAN for SISR reduces diffusion timesteps via the diffusion GAN paradigm and applies DWT for dimensionality reduction, claiming faster training/inference and higher fidelity than prior methods on CelebA-HQ.