RAP-SR: RestorAtion Prior Enhancement in Diffusion Models for Realistic Image Super-Resolution

TL;DR

RAP-SR enhances pretrained diffusion models for realistic image super-resolution by developing a new high-fidelity aesthetic dataset and a framework that refines restoration priors and optimizes prompts, leading to state-of-the-art results.

Contribution

The paper introduces RAP-SR, a novel method that improves restoration priors in pretrained diffusion models for Real-SR, including a new dataset and a refinement framework.

Findings

Achieves state-of-the-art super-resolution quality.

Effectively integrates with existing diffusion-based methods.

Demonstrates broad applicability across various models.

Abstract

Benefiting from their powerful generative capabilities, pretrained diffusion models have garnered significant attention for real-world image super-resolution (Real-SR). Existing diffusion-based SR approaches typically utilize semantic information from degraded images and restoration prompts to activate prior for producing realistic high-resolution images. However, general-purpose pretrained diffusion models, not designed for restoration tasks, often have suboptimal prior, and manually defined prompts may fail to fully exploit the generated potential. To address these limitations, we introduce RAP-SR, a novel restoration prior enhancement approach in pretrained diffusion models for Real-SR. First, we develop the High-Fidelity Aesthetic Image Dataset (HFAID), curated through a Quality-Driven Aesthetic Image Selection Pipeline (QDAISP). Our dataset not only surpasses existing ones in fidelity but also excels in aesthetic quality. Second, we propose the Restoration Priors Enhancement Framework, which includes Restoration Priors Refinement (RPR) and Restoration-Oriented Prompt Optimization (ROPO) modules. RPR refines the restoration prior using the HFAID, while ROPO optimizes the unique restoration identifier, improving the quality of the resulting images. RAP-SR effectively bridges the gap between general-purpose models and the demands of Real-SR by enhancing restoration prior. Leveraging the plug-and-play nature of RAP-SR, our approach can be seamlessly integrated into existing diffusion-based SR methods, boosting their performance. Extensive experiments demonstrate its broad applicability and state-of-the-art results. Codes and datasets will be available upon acceptance.