https://github.com/denofequity/forge_space_invsr
Forge Space for InvSR: Arbitrary-steps Image Super-resolution via Diffusion Inversion
https://github.com/denofequity/forge_space_invsr
image-super-resolution image-to-image invsr superresolution
Last synced: about 1 month ago
JSON representation
Forge Space for InvSR: Arbitrary-steps Image Super-resolution via Diffusion Inversion
- Host: GitHub
- URL: https://github.com/denofequity/forge_space_invsr
- Owner: DenOfEquity
- Created: 2025-01-06T23:10:30.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-09-26T12:33:41.000Z (10 months ago)
- Last Synced: 2025-09-26T14:31:37.815Z (10 months ago)
- Topics: image-super-resolution, image-to-image, invsr, superresolution
- Language: Python
- Homepage:
- Size: 162 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## Forge2 Spaces implementation of InvSR 4x super resolution ##
### Arbitrary-steps Image Super-resolution via Diffusion Inversion ###
New Forge only.
derived from:
* https://github.com/zsyOAOA/InvSR
* https://huggingface.co/spaces/OAOA/InvSR
[LICENSE: S-LAB 1.0](https://github.com/zsyOAOA/InvSR/blob/master/LICENSE)
will download ~4.9GB of models
>[!NOTE]
>Install via *Extensions* tab; *Install from URL* sub-tab; use URL of this repo
>
>Requirements: (edit `requirements_versions.txt` in the webUI folder)
>
>`accelerate==0.26.0`
>
>`diffusers>=0.32.0`
>[!IMPORTANT]
>You need to copy/save the result manually.