Realism Control One-step Diffusion for Real-World Image Super-Resolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Zongliang, Zheng, Siming, Jiang, Peng-Tao, Yuan, Xin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911267463102464
author Wu, Zongliang
Zheng, Siming
Jiang, Peng-Tao
Yuan, Xin
author_facet Wu, Zongliang
Zheng, Siming
Jiang, Peng-Tao
Yuan, Xin
contents Pre-trained diffusion models have shown great potential in real-world image super-resolution (Real-ISR) tasks by enabling high-resolution reconstructions. While one-step diffusion (OSD) methods significantly improve efficiency compared to traditional multi-step approaches, they still have limitations in balancing fidelity and realism across diverse scenarios. Since the OSDs for SR are usually trained or distilled by a single timestep, they lack flexible control mechanisms to adaptively prioritize these competing objectives, which are inherently manageable in multi-step methods through adjusting sampling steps. To address this challenge, we propose a Realism Controlled One-step Diffusion (RCOD) framework for Real-ISR. RCOD provides a latent domain grouping strategy that enables explicit control over fidelity-realism trade-offs during the noise prediction phase with minimal training paradigm modifications and original training data. A degradation-aware sampling strategy is also introduced to align distillation regularization with the grouping strategy and enhance the controlling of trade-offs. Moreover, a visual prompt injection module is used to replace conventional text prompts with degradation-aware visual tokens, enhancing both restoration accuracy and semantic consistency. Our method achieves superior fidelity and perceptual quality while maintaining computational efficiency. Extensive experiments demonstrate that RCOD outperforms state-of-the-art OSD methods in both quantitative metrics and visual qualities, with flexible realism control capabilities in the inference stage.
format Preprint
id arxiv_https___arxiv_org_abs_2509_10122
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Realism Control One-step Diffusion for Real-World Image Super-Resolution
Wu, Zongliang
Zheng, Siming
Jiang, Peng-Tao
Yuan, Xin
Computer Vision and Pattern Recognition
Artificial Intelligence
Pre-trained diffusion models have shown great potential in real-world image super-resolution (Real-ISR) tasks by enabling high-resolution reconstructions. While one-step diffusion (OSD) methods significantly improve efficiency compared to traditional multi-step approaches, they still have limitations in balancing fidelity and realism across diverse scenarios. Since the OSDs for SR are usually trained or distilled by a single timestep, they lack flexible control mechanisms to adaptively prioritize these competing objectives, which are inherently manageable in multi-step methods through adjusting sampling steps. To address this challenge, we propose a Realism Controlled One-step Diffusion (RCOD) framework for Real-ISR. RCOD provides a latent domain grouping strategy that enables explicit control over fidelity-realism trade-offs during the noise prediction phase with minimal training paradigm modifications and original training data. A degradation-aware sampling strategy is also introduced to align distillation regularization with the grouping strategy and enhance the controlling of trade-offs. Moreover, a visual prompt injection module is used to replace conventional text prompts with degradation-aware visual tokens, enhancing both restoration accuracy and semantic consistency. Our method achieves superior fidelity and perceptual quality while maintaining computational efficiency. Extensive experiments demonstrate that RCOD outperforms state-of-the-art OSD methods in both quantitative metrics and visual qualities, with flexible realism control capabilities in the inference stage.
title Realism Control One-step Diffusion for Real-World Image Super-Resolution
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2509.10122