SPA

This repo. is the official implementation of 'SPA: Self-Peripheral-Attention for Central-Peripheral Interactions in Endoscopic Image Classification and Segmentation'.
Authors: Xiangzuo Huo, Shengwei Tian, Yongxu Yang, Long Yu, Wendong Zhang, Aolun Li.
Enjoy the code and find its convenience to produce more awesome works!

Motivation

SPA-Net

Self-Peripheral-Attention

Visual Inspection of SPA

Segmentation Results

Run

0. Requirements:

• python3
• pytorch 1.12.0
• torchvision 0.13.0

1. Train:

• Prepare the required images and store them in categories, set up training image folders and validation image folders respectively
• Run python train.py

2. Resume train:

• Modify parser.add_argument('--RESUME', type=bool, default=True) in python train.py
• Run python train.py

3. Test:

• Run python test.py

4. Predict:

• Run python predict.py

TensorBoard

Run tensorboard --logdir runs --port 6006 to view training progress

Reference

Some of the codes in this repo are borrowed from:

• SwinTransformer
• FocalNet
• WZMIAOMIAO

Citation

If you find our paper/code is helpful, please consider citing:

@article{huo2024spa,
  title={SPA: Self-Peripheral-Attention for central--peripheral interactions in endoscopic image classification and segmentation},
  author={Huo, Xiangzuo and Tian, Shengwei and Yang, Yongxu and Yu, Long and Zhang, Wendong and Li, Aolun},
  journal={Expert Systems with Applications},
  volume={245},
  pages={123053},
  year={2024},
  publisher={Elsevier}
}