DreamClear: High-Capacity Real-World Image Restoration with Privacy-Safe Dataset Curation

TL;DR

DreamClear introduces a large-scale, privacy-safe dataset creation pipeline and a novel diffusion transformer-based model with adaptive modulation, significantly advancing real-world image restoration capabilities.

Contribution

The paper presents GenIR, a dual-prompt data curation pipeline for large-scale datasets, and DreamClear, a diffusion transformer model with adaptive modulation for improved real-world IR.

Findings

DreamClear outperforms existing IR models in real-world scenarios.

GenIR successfully creates a one-million-image dataset with privacy safety.

The MoAM module enhances the model's ability to handle diverse degradations.

Abstract

Image restoration (IR) in real-world scenarios presents significant challenges due to the lack of high-capacity models and comprehensive datasets. To tackle these issues, we present a dual strategy: GenIR, an innovative data curation pipeline, and DreamClear, a cutting-edge Diffusion Transformer (DiT)-based image restoration model. GenIR, our pioneering contribution, is a dual-prompt learning pipeline that overcomes the limitations of existing datasets, which typically comprise only a few thousand images and thus offer limited generalizability for larger models. GenIR streamlines the process into three stages: image-text pair construction, dual-prompt based fine-tuning, and data generation & filtering. This approach circumvents the laborious data crawling process, ensuring copyright compliance and providing a cost-effective, privacy-safe solution for IR dataset construction. The result is a large-scale dataset of one million high-quality images. Our second contribution, DreamClear, is a DiT-based image restoration model. It utilizes the generative priors of text-to-image (T2I) diffusion models and the robust perceptual capabilities of multi-modal large language models (MLLMs) to achieve photorealistic restoration. To boost the model's adaptability to diverse real-world degradations, we introduce the Mixture of Adaptive Modulator (MoAM). It employs token-wise degradation priors to dynamically integrate various restoration experts, thereby expanding the range of degradations the model can address. Our exhaustive experiments confirm DreamClear's superior performance, underlining the efficacy of our dual strategy for real-world image restoration. Code and pre-trained models are available at: https://github.com/shallowdream204/DreamClear.