AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Korkmaz, Cansu, Mehta, Nancy, Timofte, Radu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929751788093440
author Korkmaz, Cansu
Mehta, Nancy
Timofte, Radu
author_facet Korkmaz, Cansu
Mehta, Nancy
Timofte, Radu
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2503_07748
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution
Korkmaz, Cansu
Mehta, Nancy
Timofte, Radu
Image and Video Processing
Computer Vision and Pattern Recognition
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.
title AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.07748