FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis

arXiv:2403.12963v1 Announce Type: new
Abstract: In this study, we delve into the generation of high-resolution images from pre-trained diffusion models, addressing persistent challenges, such as repetitive patterns and structural distortions, that emerge when models are applied beyond their trained resolutions. To address this issue, we introduce an innovative, training-free approach FouriScale from the perspective of frequency domain analysis. We replace the original convolutional layers in pre-trained diffusion models by incorporating a dilation technique along with a low-pass operation, intending to …

arxiv cs.cv free image perspective synthesis training type