AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution
Preprint
Abstract
Recovering high-frequency details and textures from low-resolution images remains a fundamental challenge in super-resolution (SR), especially when real-world degradations are complex and unknown. While GAN-based methods enhance realism, they suffer from training instability and introduce unnatural artifacts. Diffusion models, though promising, demand excessive computational resources, often requiring multiple GPU days, even for single-step variants. Rather than naively fine-tuning entire models or adopting unstable generative approaches, we introduce AdaptSR, a low-rank adaptation (LoRA) framework that efficiently repurposes bicubic-trained SR models for real-world tasks. AdaptSR leverages architecture-specific insights and selective layer updates to optimize real SR adaptation. By updating only lightweight LoRA layers while keeping the pretrained backbone intact, it captures domain-specific adjustments without adding inference cost, as the adapted layers merge seamlessly post-training. This efficient adaptation not only reduces memory and compute requirements but also makes real-world SR feasible on lightweight hardware. Our experiments demonstrate that AdaptSR outperforms GAN and diffusion-based SR methods by up to 4 dB in PSNR and 2% in perceptual scores on real SR benchmarks. More impressively, it matches or exceeds full model fine-tuning while training 92% fewer parameters, enabling rapid adaptation to real SR tasks within minutes.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Large Language Models for Software Engineering: A Systematic Literature ReviewACM Transactions on Software Engineering and Methodology, 2024
- Contribution-Based Low-Rank Adaptation with Pre-training Model for Real Image Restoration2024
- Roadmap on computational methods in optical imaging and holography [invited]Applied Physics B, 2024
- Exploiting Diffusion Prior for Real-World Image Super-ResolutionInternational Journal of Computer Vision, 2024
- Super‐Resolution Two‐Photon Fluorescence Tomography Through the Phase‐Shifted Optical Beatings of Bessel Beams for High‐Resolution Deeper Tissue 3D ImagingLaser & Photonics Reviews, 2023
- Medical image super-resolution reconstruction algorithms based on deep learning: A surveyComputer Methods and Programs in Biomedicine, 2023
- Large-scale individual building extraction from open-source satellite imagery via super-resolution-based instance segmentation approachISPRS Journal of Photogrammetry and Remote Sensing, 2022
- Using Super-Resolution Algorithms for Small Satellite Imagery: A Systematic ReviewIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022
- Cross Aggregation Transformer for Image Restoration2022
- Image Quality Assessment: Unifying Structure and Texture SimilarityIEEE Transactions on Pattern Analysis and Machine Intelligence, 2020