CSCF-Net

CASF-Net: Cross-attention And Cross-scale Fusion Network for Medical Image Segmentation (Submitted)

CASF-Net Architecture

1. Environment setup

This code has been tested on on a personal laptop with Intel i7-10700H 3.8-GHz processor, 32-GB RAM, and an NVIDIA GTX3060t graphic card, Python 3.6, PyTorch 9.1, CUDA 10.1, cuDNN 8.3.

2. Downloading necessary data:

• Kvasir-SEG Dataset: Segmented Polyp Dataset for Computer Aided Gastrointestinal Disease Detection

• ISIC 2018: ISIC 2018 Challenge

• GLAS,: Gland Segmentation Challenge Contest

and put them into data directory.

3. Download Related Documents

the models (loading models) ， pretrained_models (loading model parameters) and result map:

• Related Documents

and put them into models directory.

4.Running

python  XX.py

The result will be saved in snapshots...

5. Reference

Some of the codes in this repo are borrowed from:

timm repo

PraNet repo

TransFuse repo

6. Questions

Please drop an email to hiderhao@gmail.com