cjwbw / supir

Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild. This version uses LLaVA-13b for captioning.

  • Public
  • 155K runs
  • GitHub
  • Paper
  • License
Run with an API
Playground API Examples README Versions

Examples

View more

Run time and cost

This model costs approximately $0.065 to run on Replicate, or 15 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia A100 (80GB) GPU hardware. Predictions typically complete within 47 seconds. The predict time for this model varies significantly based on the inputs.

Readme

Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild

NOTE: Due to the memory requirement for LLaVA-13b and both SUPIR-v0F and SUPIR-v0Q, this model is posted on 80G A100. To run on 40G A40, try models https://replicate.com/cjwbw/supir-v0q or https://replicate.com/cjwbw/supir-v0f that do not include LLaVA-13b.

  • SUPIR-v0Q: Default training settings with paper. High generalization and high image quality in most cases.
  • SUPIR-v0F: Training with light degradation settings. Stage1 encoder of SUPIR-v0F remains more details when facing light degradations.

BibTeX

@misc{yu2024scaling,
  title={Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild}, 
  author={Fanghua Yu and Jinjin Gu and Zheyuan Li and Jinfan Hu and Xiangtao Kong and Xintao Wang and Jingwen He and Yu Qiao and Chao Dong},
  year={2024},
  eprint={2401.13627},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}