Dual-domain Adaptation Networks for Realistic Image Super-resolution

2025-11-21

AI Summary

In real-world image super-resolution, the scarcity of paired low-resolution (LR)–high-resolution (HR) data and complex, unknown degradation patterns severely hinder model generalization. To address this, we propose the Dual-Domain Adaptive Network (DDAN), a novel framework integrating adaptive mechanisms in both spatial and frequency domains. Spatially, DDAN employs selective parameter updating combined with Low-Rank Adaptation (LoRA) for lightweight, efficient domain transfer. Frequency-wise, it introduces a spectral-aware branch that jointly fuses input spectral information and intermediate features to enhance high-frequency detail reconstruction. By jointly optimizing a synthetically pre-trained model toward real-world domains, DDAN achieves state-of-the-art performance on RealSR, D2CRealSR, and DRealSR benchmarks—yielding substantial PSNR and SSIM improvements over prior methods and producing visually sharper, more natural reconstructions.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Diffusion Models for Vision
• Application Domains: Software Engineering

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Realistic image super-resolution (SR) focuses on transforming real-world low-resolution (LR) images into high-resolution (HR) ones, handling more complex degradation patterns than synthetic SR tasks. This is critical for applications like surveillance, medical imaging, and consumer electronics. However, current methods struggle with limited real-world LR-HR data, impacting the learning of basic image features. Pre-trained SR models from large-scale synthetic datasets offer valuable prior knowledge, which can improve generalization, speed up training, and reduce the need for extensive real-world data in realistic SR tasks. In this paper, we introduce a novel approach, Dual-domain Adaptation Networks, which is able to efficiently adapt pre-trained image SR models from simulated to real-world datasets. To achieve this target, we first set up a spatial-domain adaptation strategy through selectively updating parameters of pre-trained models and employing the low-rank adaptation technique to adjust frozen parameters. Recognizing that image super-resolution involves recovering high-frequency components, we further integrate a frequency domain adaptation branch into the adapted model, which combines the spectral data of the input and the spatial-domain backbone's intermediate features to infer HR frequency maps, enhancing the SR result. Experimental evaluations on public realistic image SR benchmarks, including RealSR, D2CRealSR, and DRealSR, demonstrate the superiority of our proposed method over existing state-of-the-art models. Codes are available at: https://github.com/dummerchen/DAN.

Problem

• Adapting pre-trained SR models from synthetic to real-world datasets
• Handling complex degradation patterns in realistic image super-resolution
• Overcoming limited real-world LR-HR data through dual-domain adaptation

Innovation

• Adapts pre-trained SR models to real-world datasets
• Uses spatial-domain adaptation with selective parameter updates
• Integrates frequency domain branch for enhanced HR recovery