[CVPR 2026 Findings] SAT

June 1, 2026 · View on GitHub

[CVPR 2026 Findings] Official repository for "SAT: Selective Aggregation Transformer for Image Super-Resolution"

Dinh Phu Tran, Thao Do, Saad Wazir, Seongah Kim, Seon Kwon Kim and Daeyoung Kim

arXiv visitors GitHub Stars

Updates

  • ✅ 2026-04-01: Release repository

Overview

Abstract:

Transformer-based approaches have revolutionized image super-resolution by modeling long-range dependencies. However, the quadratic computational complexity of vanilla self-attention mechanisms poses significant challenges, often leading to compromises between efficiency and global context exploitation. Recent window-based attention methods mitigate this by localizing computations, but they often yield restricted receptive fields. To mitigate these limitations, we propose Selective Aggregation Transformer (SAT). This novel transformer efficiently captures long-range dependencies, leading to an enlarged model receptive field by selectively aggregating key-value matrices (reducing the number of tokens by 97%) via our Density-driven Token Aggregation algorithm while maintaining the full resolution of the query matrix. This design significantly reduces computational costs, resulting in lower complexity and enabling scalable global interactions without compromising reconstruction fidelity. SAT identifies and represents each cluster with a single aggregation token, utilizing density and isolation metrics to ensure that critical high-frequency details are preserved. Experimental results demonstrate that SAT outperforms the state-of-the-art method PFT by up to 0.22dB, while the total number of FLOPs can be reduced by up to 27%.

Overall Architecture of SAT:

Selective Aggregation Attention:

Quantitative Results:

Qualitative Results:

Environment

Installation

Install Pytorch first. Then,

pip install -r requirements.txt
python setup.py develop

How To Test

  • Run the following codes:
python basicsr/test.py -opt options/test/SAT_SRx4.yml

The testing results will be saved in the ./results folder.

How To Train

  • Refer to ./options/train for the configuration file of the model to train.

  • Preparation of training data can refer to this page.

  • The training command is like

CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node=4 --master_port=4321 basicsr/train.py -opt options/train/train_SAT_SRx4.yml --launcher pytorch

The training logs and weights will be saved in the ./experiments folder.

Citations

BibTeX

TBD

Acknowledgement

Our codes was built on BasicSR and RGT. Thank you authors for sharing their great works.