This repository is an official Pytorch implementation of the paper "Clustering-Based Adaptive Query Generation for Semantic Segmentation"
Yeong Woo Kim and Wonjun Kim
IEEE Signal Processing Letters, 2025.
The overall architecture of the proposed method.
Since this code is based on Mask2Former, please follow the Installation Guide of Mask2Former.
Since this code is based on Mask2Former, please follow the Dataset Preparation Guide of Mask2Former.
sh train_ade.sh # for the ADE20K experiment
sh train_city.sh # for the Cityscapes experiment| Model | Dataset | Backbone | mIoU |
|---|---|---|---|
| CQG | ADE20K | R50 | 48.7 |
| CQG | ADE20K | Swin-B | 55.4 |
| CQG | ADE20K | Swin-L | 56.4 |
| CQG | Cistyscaes | R50 | 80.8 |
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government [Ministry of Science and ICT (MSIT)] under Grant RS-2023-NR076462.
Our implementation and experiments are built on top of open-source GitHub repositories. We thank all the authors who made their code public, which tremendously accelerates our project progress. If you find these works helpful, please consider citing them as well.
[Mask2Former] https://github.com/facebookresearch/Mask2Former
@ARTICLE{10949765,
author={Kim, Yeong Woo and Kim, Wonjun},
journal={IEEE Signal Processing Letters},
title={Clustering-Based Adaptive Query Generation for Semantic Segmentation},
year={2025},
volume={32},
number={},
pages={1580-1584},
doi={10.1109/LSP.2025.3558160}}