MFSR: Multi-fractal Feature for Super-resolution Reconstruction with Fine Details Recovery

2025-02-27

AI Summary

Addressing the challenge of recovering complex local textures—particularly edges and fine structures—in image super-resolution, this paper proposes a diffusion-based multi-fractal feature-guided method. The approach models the fractal characteristics of low-resolution images as a conditional prior for the diffusion denoising process—a novel formulation. It introduces soft-assignment convolution to approximate fractal dimension and density features, enabling hierarchical encoding of cross-scale self-similarity. Additionally, a lightweight sub-denoiser is incorporated to refine upsampled feature maps with enhanced noise suppression. Integrated within a U-Net backbone, the fractal-guided mechanism achieves significant PSNR/SSIM improvements on face and natural image benchmarks. Qualitatively, it outperforms state-of-the-art methods in texture sharpness, edge clarity, and fidelity of microscopic structures.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multimodal Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

In the process of performing image super-resolution processing, the processing of complex localized information can have a significant impact on the quality of the image generated. Fractal features can capture the rich details of both micro and macro texture structures in an image. Therefore, we propose a diffusion model-based super-resolution method incorporating fractal features of low-resolution images, named MFSR. MFSR leverages these fractal features as reinforcement conditions in the denoising process of the diffusion model to ensure accurate recovery of texture information. MFSR employs convolution as a soft assignment to approximate the fractal features of low-resolution images. This approach is also used to approximate the density feature maps of these images. By using soft assignment, the spatial layout of the image is described hierarchically, encoding the self-similarity properties of the image at different scales. Different processing methods are applied to various types of features to enrich the information acquired by the model. In addition, a sub-denoiser is integrated in the denoising U-Net to reduce the noise in the feature maps during the up-sampling process in order to improve the quality of the generated images. Experiments conducted on various face and natural image datasets demonstrate that MFSR can generate higher quality images.

Problem

• Enhance image super-resolution quality
• Recover fine details using fractal features
• Reduce noise in up-sampling process

Innovation

• Diffusion model with fractal features
• Soft assignment for feature approximation
• Sub-denoiser in denoising U-Net