The Year in AI Papers
The year/Independent research

Paper 2603.25502

RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models

Yufeng Yang, Xianfang Zeng, Zhangqi Jiang, Fukun Yin, Jianzhuang Liu, Wei Cheng, jinghong lan, Shiyu Liu, Yuqi Peng, Gang YU, Shifeng Chen

Published
Mar 2026
Research lab
Independent
Citations
3
GitHub
Not linked

01 In brief

Summary

RealRestorer is an open-source image restoration model designed to handle diverse real-world degradations, including blur, rain, noise, low-light, moiré patterns, haze, compression artifacts, reflection, and flare.

The authors construct a large-scale dataset with a synthesis pipeline that combines synthetic and real-world degradation data, and they introduce RealIR-Bench, a benchmark of 464 real-world degraded images with non-reference evaluation metrics (Restoration Score and LPIPS-based Final Score).

The model is fine-tuned from Step1X-Edit using a two-stage training strategy: transfer training on synthetic data followed by supervised fine-tuning with real-world data, incorporating a progressively-mixed approach.

Experiments show RealRestorer ranks first among open-source methods and is competitive with closed-source systems like Nano Banana Pro, achieving state-of-the-art performance on tasks such as deblurring and low-light enhancement.

The model also demonstrates zero-shot generalization to unseen tasks like snow removal and old photo restoration.

The authors will release the model, data pipeline, and benchmark to support future research.

02 From the paper

Abstract

Image restoration under real-world degradations is critical for downstream tasks such as autonomous driving and object detection. However, existing restoration models are often limited by the scale and distribution of their training data, resulting in poor generalization to real-world scenarios. Recently, large-scale image editing models have shown strong generalization ability in restoration tasks, especially for closed-source models like Nano Banana Pro, which can restore images while preserving consistency. Nevertheless, achieving such performance with those large universal models requires substantial data and computational costs. To address this issue, we construct a large-scale dataset covering nine common real-world degradation types and train a state-of-the-art open-source model to narrow the gap with closed-source alternatives. Furthermore, we introduce RealIR-Bench, which contains 464 real-world degraded images and tailored evaluation metrics focusing on degradation removal and consistency preservation. Extensive experiments demonstrate our model ranks first among open-source methods, achieving state-of-the-art performance.