# swz30/MPRNet

[CVPR 2021] Multi-Stage Progressive Image Restoration. SOTA results for Image deblurring, deraining, and denoising.

Repository: https://github.com/swz30/MPRNet
Canonical: https://ross.abutalabs.com/products/mprnet
Language: Python
License: NOASSERTION
License Family: other
Topics: image-deblurring, image-denoising, image-deraining, low-level-vision, computer-vision, image-restoration, multistage-network, progressive-restoration, pytorch, cvpr2021, cvpr21, cvpr-2021
Last push: 2024-04-18T08:04:26+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2036, "days_push": 867, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1393, forks 207 (observed 2026-08-28T04:04:36.271351+00:00)

## What it is
MPRNet is a PyTorch implementation of a multi-stage progressive image restoration network published at CVPR 2021. It provides pretrained models and training code for state-of-the-art image deblurring, deraining, and denoising.

## Use cases
- deblur blurry photos with a deep learning model
- remove rain streaks from images
- denoise noisy photographs
- train an image restoration network in PyTorch
- reproduce CVPR 2021 image restoration results
- restore degraded images with a pretrained model

## When to choose
- you need strong deblurring, deraining, or denoising with pretrained weights
- you want a research baseline for low-level vision restoration
- you work in PyTorch and want a multi-stage restoration architecture

## When to avoid
- you need a production image-processing pipeline rather than research code
- you need general image enhancement beyond deblurring/deraining/denoising
- you require a permissive license - the license is non-standard

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, machine-learning
- platform: python, cross-platform
- tags: image-restoration, image-deblurring, image-denoising, image-deraining, pytorch, cvpr-2021, low-level-vision, research-code, gpu, linux

## Member repositories
- swz30/MPRNet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:36.271351+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 2026-08-30T04:39:30.360176+00:00, confidence not recorded.
  - readme: https://github.com/swz30/MPRNet (fetched 2026-08-28T04:04:36.271351+00:00, sha 330dce1f4ca9)
- Data as of 2026-08-30T08:39:29.467469+00:00.