# andreas128/RePaint

Official PyTorch Code and Models of "RePaint: Inpainting using Denoising Diffusion Probabilistic Models", CVPR 2022

Repository: https://github.com/andreas128/RePaint
Canonical: https://ross.abutalabs.com/products/repaint
Language: Python
License Family: other
Topics: cvpr2022, diffusion-models, inpainting
Last push: 2022-08-20T13:01:44+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": 1738, "days_push": 1474, "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 2267, forks 198 (observed 2026-08-28T04:06:32.765555+00:00)

## What it is
Official PyTorch implementation of RePaint, a CVPR 2022 paper on image inpainting using denoising diffusion probabilistic models. It provides pretrained models for ImageNet, CelebA-HQ, and Places2, filling in missing image regions guided by known pixels.

## Use cases
- fill in missing parts of images with a diffusion model
- inpaint faces or diverse photos using pretrained models
- reproduce CVPR 2022 RePaint paper results
- run image inpainting with masks like thin, thick, or half
- experiment with diffusion model resampling schedules
- super-resolution or image expansion via inpainting

## When to choose
- you want research-grade diffusion-based inpainting with pretrained checkpoints
- you need to reproduce or build on the RePaint paper
- your images match CelebA-HQ, ImageNet, or Places2 style domains

## When to avoid
- you need a production-ready tool with a maintained license
- you want a general-purpose inpainting GUI or plugin
- you need support for arbitrary domains without retraining

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: python
- tags: diffusion-models, inpainting, pytorch, cvpr2022, research-code, image-generation, gpu, linux, macos

## Member repositories
- andreas128/RePaint (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:32.765555+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 2026-08-30T02:42:06.567449+00:00, confidence not recorded.
  - readme: https://github.com/andreas128/RePaint (fetched 2026-08-28T04:06:32.765555+00:00, sha 9f6aa6f57d2f)
- Data as of 2026-08-30T08:39:29.467469+00:00.