diffusers icon indicating copy to clipboard operation
diffusers copied to clipboard

CCSR: Improving the Stability of Diffusion Models for Content Consistent Super-Resolution

Open clarencechen opened this issue 1 year ago • 3 comments
trafficstars

Model/Pipeline/Scheduler description

The generative priors of pre-trained latent diffusion models have demonstrated great potential to enhance the perceptual quality of image super-resolution (SR) results. Unfortunately, the existing diffusion prior-based SR methods encounter a common problem, i.e., they tend to generate rather different outputs for the same low-resolution image with different noise samples. Such stochasticity is desired for text-to-image generation tasks but problematic for SR tasks, where the image contents are expected to be well preserved. To improve the stability of diffusion prior-based SR, the authors propose to employ the diffusion models to refine image structures, while employing the generative adversarial training to enhance image fine details. Specifically, the authors propose a non-uniform timestep learning strategy to train a compact diffusion network, which has high efficiency and stability to reproduce the image main structures, and finetune the pre-trained decoder of variational auto-encoder (VAE) by adversarial training for detail enhancement. Extensive experiments show that our proposed method, namely content consistent super-resolution (CCSR), can significantly reduce the stochasticity of diffusion prior-based SR, improving the content consistency of SR outputs and speeding up the image generation process.

Open source status

  • [X] The model implementation is available.
  • [X] The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

  • Github: https://github.com/csslc/CCSR
  • Paper: https://arxiv.org/pdf/2401.00877.pdf
  • Project Page: https://csslc.github.io/project-CCSR/
  • First Author: @csslc

clarencechen avatar Jan 27 '24 00:01 clarencechen

It'd be pretty amazing if this model was ported to diffusers!!

isidentical avatar Feb 09 '24 21:02 isidentical

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.

github-actions[bot] avatar Mar 05 '24 15:03 github-actions[bot]

âž•

isidentical avatar Mar 05 '24 16:03 isidentical

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.

github-actions[bot] avatar Apr 02 '24 15:04 github-actions[bot]