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

TL;DR

MFSR introduces a diffusion model that uses fractal features of low-resolution images to enhance super-resolution reconstruction, especially in recovering fine textures and details.

Contribution

The paper proposes a novel super-resolution method integrating fractal features into a diffusion model, improving texture detail recovery in images.

Findings

MFSR achieves higher quality image reconstruction on face and natural image datasets.

Incorporating fractal features improves the preservation of micro and macro textures.

The method outperforms existing super-resolution techniques in detail recovery.

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.