---
license: apache-2.0
pipeline_tag: image-to-image
---

# S3Diff Model Card
This model card focuses on the models associated with the S3Diff, available [here](https://github.com/ArcticHare105/S3Diff).

## Model Details
- **Developed by:** Aiping Zhang
- **Model type:** Degradation-Guided One-Step Image Super-Resolution with Diffusion Priors
- **Model Description:** This is the model used in [Paper](https://arxiv.org/abs/2409.17058).
- **Resources for more information:** [GitHub Repository](https://github.com/ArcticHare105/S3Diff).
- **Cite as:**

      @article{2024s3diff,
        author    = {Aiping Zhang, Zongsheng Yue, Renjing Pei, Wenqi Ren, Xiaochun Cao},
        title     = {Degradation-Guided One-Step Image Super-Resolution with Diffusion Priors},
        journal   = {arxiv},
        year      = {2024},
      }

## Limitations and Bias

### Limitations

- S3Diff requires a tiled operation for generating a high-resolution image, which would largely increase the inference time.
- S3Diff sometimes cannot keep 100% fidelity due to its generative nature.
- S3Diff sometimes cannot generate perfect details under complex real-world scenarios.

### Bias
While our model is based on a pre-trained SD-Turbo model, currently we do not observe obvious bias in generated results.
We conjecture the main reason is that our model does not rely on text prompts but on low-resolution images.
Such strong conditions make our model less likely to be affected.

## Training

**Training Data**
The model developer used the following dataset for training the model:

- Our model is finetuned on [LSDIR](https://data.vision.ee.ethz.ch/yawli/index.html) + 10K samples from FFHQ datasets.

**Training Procedure**
S3Diff is an image super-resolution model finetuned on [SD-Turbo](https://huggingface.co/stabilityai/sd-turbo), further equipped with a degradation-guided LoRA and online negative prompting.

- Following SD-Turbo, images are encoded through the fixed autoencoder, which turns images into latent representations. The autoencoder uses a relative downsampling factor of 8 and maps images of shape H x W x 3 to latents of shape H/f x W/f x 4.
- The LR images are fed to the degradation estimation network, trained by [mm-realsr](https://github.com/TencentARC/MM-RealSR), to predict degradation scores.
- We only inject LoRA layers into the VAE encoder and UNet.
- The total loss includes an L2 Loss, an LPIPS loss, and a GAN loss.

We currently provide the following checkpoints:

- [s3diff.pkl](https://huggingface.co/zhangap/S3Diff/blob/main/s3diff.pkl): S3Diff finetuned on [SD-Turbo](https://huggingface.co/stabilityai/sd-turbo) for 30k iterations.
- [de_net.pth](https://huggingface.co/zhangap/S3Diff/blob/main/de_net.pth): The degradation estimation network, extracted from [mm-realsr](https://github.com/TencentARC/MM-RealSR).

## Evaluation Results 
See [Paper](https://arxiv.org/abs/2409.17058) for details.