Skip to content

XUSean0118/FPLIA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Yu-Syuan Xu , Hao-Lun Sun , Hao-Wei Chen , Hsien-Kai Kuo , Chun-Yi Lee

🎉 Paper accepted at ECCV 2026


Abstract: Arbitrary-scale image super-resolution (ASISR) aims to reconstruct high-resolution images from low-resolution inputs over a continuous range of upscaling factors. While traditional pixel-regression approaches often produce overly smooth results that lack realistic details, recent diffusion methods can produce sharper and more realistic textures. However, these diffusion techniques frequently introduce the risk of structural hallucinations. To address these issues, we propose Fidelity- and Perception-Aware Local Implicit Attention (FPLIA), a framework that effectively integrates fidelity-oriented features into a diffusion pipeline to produce realistic and faithful reconstructions for ASISR. We introduce a Fidelity and Perception Attention Module (FPAM), which applies both self-attention and cross-attention to fidelity-oriented and perceptual features to enhance representational capacity. To further exploit their complements, we design a Fidelity and Perception Select Module (FPSM) that adaptively selects the most representative features for RGB values prediction. We conduct extensive experiments to validate the effectiveness of these components. Both qualitative and quantitative results show that FPLIA delivers superior perceptual realism while maintaining reconstruction accuracy on standard ASISR benchmarks.



📚 Table of Contents


🏗️ Framework

Overall Framework of FPLIA


💻 Installation

We use Python=3.12.2 and PyTorch>=2.8 with single B200.

Installation of required packages:

pip install -r requirements.txt

🚂 Training

python3 main.py --base [CONFIG] --device [NUM_GPU]
Parameter Description
CONFIG Path to the configuration file.
NUM_GPU Number of GPUs.

🧪 Testing

[>> Checkpoints <<]

python3 test.py --exp [EXP_FOLDER] --dataset [DATASET] (--save_image)
Parameter Description
EXP_FOLDER Path to the trained experiment folder.
DATASET Name of the dataset.
--save_image Save the output images. (Optional).

📜 Acknowledgements

This repo is built on CLIT and arbitrary-scale-diffusion. Thanks the authors for their contributions and generosity.


About

Official repo for FPLIA: Fidelity- and Perception-Aware Local Implicit Attention for Arbitrary-Scale Image Super-Resolution.

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages