CVMay 8, 2025

EAM: Enhancing Anything with Diffusion Transformers for Blind Super-Resolution

arXiv:2505.05209v3h-index: 5
Originality Highly original
AI Analysis

This work addresses image restoration for applications like photography and medical imaging, representing an incremental improvement over existing methods.

The paper tackles blind super-resolution by introducing the Enhancing Anything Model (EAM), which leverages Diffusion Transformers and novel components like Ψ-DiT and a progressive Masked Image Modeling strategy, achieving state-of-the-art results across multiple datasets.

Utilizing pre-trained Text-to-Image (T2I) diffusion models to guide Blind Super-Resolution (BSR) has become a predominant approach in the field. While T2I models have traditionally relied on U-Net architectures, recent advancements have demonstrated that Diffusion Transformers (DiT) achieve significantly higher performance in this domain. In this work, we introduce Enhancing Anything Model (EAM), a novel BSR method that leverages DiT and outperforms previous U-Net-based approaches. We introduce a novel block, $Ψ$-DiT, which effectively guides the DiT to enhance image restoration. This block employs a low-resolution latent as a separable flow injection control, forming a triple-flow architecture that effectively leverages the prior knowledge embedded in the pre-trained DiT. To fully exploit the prior guidance capabilities of T2I models and enhance their generalization in BSR, we introduce a progressive Masked Image Modeling strategy, which also reduces training costs. Additionally, we propose a subject-aware prompt generation strategy that employs a robust multi-modal model in an in-context learning framework. This strategy automatically identifies key image areas, provides detailed descriptions, and optimizes the utilization of T2I diffusion priors. Our experiments demonstrate that EAM achieves state-of-the-art results across multiple datasets, outperforming existing methods in both quantitative metrics and visual quality.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes