# XPixelGroup/DiffBIR

[ECCV 2024] codes of DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior

Repository: https://github.com/XPixelGroup/DiffBIR
Canonical: https://ross.abutalabs.com/products/diffbir
Language: Python
License: Apache-2.0
License Family: permissive
Topics: image-enhancement, image-restoration
Last push: 2025-07-29T12:11:31+00:00

## Health v2 (maintenance only)
Score: 34/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 34, release rhythm 8, longevity 78
- inputs: {"age_days": 1101, "days_push": 400, "days_rel": 644, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4119, forks 364 (observed 2026-08-28T04:08:35.916165+00:00)

## What it is
DiffBIR is a blind image restoration framework that uses generative diffusion priors to restore degraded real-world images. It provides pretrained models and a unified pipeline for tasks like super-resolution, deblurring, and face restoration.

## Use cases
- restore old blurry photos
- upscale low-resolution images
- fix degraded real-world images
- enhance faces in low-quality photos
- remove compression artifacts from images
- research on diffusion-based image restoration

## When to choose
- you need state-of-the-art blind image restoration with generative detail
- you want pretrained models for super-resolution or deblurring
- you are researching diffusion priors for restoration

## When to avoid
- you need fast inference, since diffusion sampling is slow (consider HYPIR)
- you only need simple classical denoising without generative detail
- you cannot run GPU inference

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: image-processing, computer-vision, artificial-intelligence
- platform: python, cross-platform
- tags: image-restoration, image-enhancement, diffusion-models, blind-restoration, super-resolution, eccv-2024, gpu

## Member repositories
- XPixelGroup/DiffBIR (main) score 34

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:35.916165+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 2026-08-29T18:23:07.584300+00:00, confidence not recorded.
  - readme: https://github.com/XPixelGroup/DiffBIR (fetched 2026-08-28T04:08:35.916165+00:00, sha 23c7ebc7f422)
- Data as of 2026-08-30T08:39:29.467469+00:00.