Official Pytorch implementation of the paper MK-UNet: Multi-kernel Lightweight CNN for Medical Image Segmentation published in ICCV 2025 CVAMD Md Mostafijur Rahman, Radu Marculescu
The University of Texas at Austin
🔍 Check out our papers: LoMix [NeurIPS 2025], EfficientMedNeXt [MICCAI 2025], EffiDec3D [CVPR 2025], EMCAD [CVPR 2024], PP-SAM [CVPRW 2024], G-CASCADE [WACV 2024], MERIT [MIDL 2023], CASCADE [WACV 2023]
Please run the following commands.
conda create -n mkunetenv python=3.8
conda activate mkunetenv
pip install torch==1.11.0+cu113 torchvision==0.12.0+cu113 torchaudio==0.11.0 --extra-index-url https://download.pytorch.org/whl/cu113
pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.11.0/index.html
pip install -r requirements.txt
-
ClinicDB dataset: Download the splited ClinicDB dataset from Google Drive and move into './data/polyp/' folder.
-
ColonDB dataset: Download the splited ColonDB dataset from Google Drive and move into './data/polyp/' folder.
cd into MK-UNet
CUDA_VISIBLE_DEVICES=0 python -W ignore train_polyp.py --network MK_UNet
cd into MK-UNet
CUDA_VISIBLE_DEVICES=0 python -W ignore test_polyp.py --network MK_UNet --run_id <your run_id>
We are very grateful for these excellent works EMCAD, CASCADE, MERIT, G-CASCADE, PP-SAM, PraNet, and Polyp-PVT, which have provided the basis for our framework.
@inproceedings{rahman2025mk,
title={Mk-unet: Multi-kernel lightweight cnn for medical image segmentation},
author={Rahman, Md Mostafijur and Marculescu, Radu},
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
pages={1042--1051},
year={2025}
}