# IceClear/StableSR

[IJCV2024] Exploiting Diffusion Prior for Real-World Image Super-Resolution

Repository: https://github.com/IceClear/StableSR
Canonical: https://ross.abutalabs.com/products/stablesr
Homepage: https://iceclear.github.io/projects/stablesr/
Language: Python
License: NOASSERTION
License Family: other
Topics: stable-diffusion, super-resolution, stablesr
Last push: 2024-07-12T03:13:49+00:00

## Health v2 (maintenance only)
Score: 21/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 89
- inputs: {"age_days": 1249, "days_push": 782, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2668, forks 171 (observed 2026-08-28T04:07:09.375231+00:00)

## What it is
StableSR is a Python research library that leverages pre-trained Stable Diffusion priors for real-world blind image super-resolution. It provides a time-aware encoder, controllable feature wrapping for quality-fidelity tradeoff, and progressive aggregation sampling for arbitrary-size upscaling.

## Use cases
- upscale low-quality real-world photos
- restore degraded images with diffusion models
- super-resolve old or compressed images
- enhance faces in low-resolution images
- run image super-resolution with stable diffusion prior
- balance fidelity and quality when upscaling

## When to choose
- you need state-of-the-art blind super-resolution on real-world degraded images
- you want to leverage Stable Diffusion priors for restoration research
- you need arbitrary-resolution upscaling with controllable quality-fidelity tradeoff

## When to avoid
- you need a lightweight fast upscaler without GPU or diffusion model overhead
- you only need simple bicubic-style resizing
- you need a production service with a stable license (license is non-standard)

## Facets
- artifact type: library
- maturity: stable
- function: image-processing, machine-learning, deep-learning, stable-diffusion
- domain: image-processing, computer-vision, deep-learning, artificial-intelligence
- platform: python, cross-platform
- tags: super-resolution, diffusion-models, stable-diffusion, image-restoration, blind-sr, research-code, gpu, linux

## Member repositories
- IceClear/StableSR (main) score 21

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:09.375231+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:18:13.089101+00:00, confidence not recorded.
  - readme: https://github.com/IceClear/StableSR (fetched 2026-08-28T04:07:09.375231+00:00, sha 83531358479a)
  - homepage: https://iceclear.github.io/projects/stablesr/ (fetched 2026-08-29T10:00:51.794659+00:00, sha bb26185eb68f)
- Data as of 2026-08-30T08:39:29.467469+00:00.