# shallowdream204/DreamClear

[NeurIPS 2024] DreamClear: High-Capacity Real-World Image Restoration with Privacy-Safe Dataset Curation

Repository: https://github.com/shallowdream204/DreamClear
Canonical: https://ross.abutalabs.com/products/dreamclear
Language: Python
License: Apache-2.0
License Family: permissive
Topics: diffusion-transformer, restoration, super-resolution, pixelart
Last push: 2025-03-21T00:49:45+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 12, release rhythm 35, longevity 48
- inputs: {"age_days": 678, "days_push": 531, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1191, forks 46 (observed 2026-08-28T04:03:56.101393+00:00)

## What it is
DreamClear is a diffusion-transformer based real-world image restoration model for high-fidelity super-resolution, published at NeurIPS 2024. It includes training and inference code, pre-trained weights, a RealLQ250 benchmark, and a privacy-safe dataset curation pipeline.

## Use cases
- restore and upscale low-quality real-world photos
- super-resolution of degraded images with diffusion models
- benchmark image restoration models on real-world LQ images
- train a diffusion transformer for image restoration
- curate privacy-safe datasets for restoration training

## When to choose
- you need state-of-the-art real-world image super-resolution
- you want pretrained restoration weights and inference code
- you are researching diffusion-based image restoration

## When to avoid
- you need fast real-time restoration on CPU
- you only need simple classical upscaling without GPU inference
- you need a production-ready service rather than research code

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, artificial-intelligence
- platform: python
- tags: super-resolution, diffusion-transformer, image-restoration, dataset-curation, neurips-2024, research-code, linux, gpu

## Member repositories
- shallowdream204/DreamClear (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:56.101393+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 2026-08-30T06:22:58.695097+00:00, confidence not recorded.
  - readme: https://github.com/shallowdream204/DreamClear (fetched 2026-08-28T04:03:56.101393+00:00, sha a96783c9e4d2)
- Data as of 2026-08-30T08:39:29.467469+00:00.