csfwinformer
Science Score: 67.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 1 DOI reference(s) in README -
✓Academic publication links
Links to: ieee.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (10.2%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: wangsen99
- License: apache-2.0
- Language: Python
- Default Branch: main
- Size: 1.6 MB
Statistics
- Stars: 6
- Watchers: 1
- Forks: 1
- Open Issues: 3
- Releases: 0
Metadata Files
README.md
CSFwinformer: Cross-Space-Frequency Window Transformer for Mirror Detection
This repo is the official implementation of "CSFwinformer: Cross-Space-Frequency Window Transformer for Mirror Detection (IEEE TIP 2024)".
Installation
``` conda create -n md python=3.7 -y conda activate md
conda install pytorch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2 cudatoolkit=11.0 -c pytorch pip install mmcv-full==1.4.0 -f https://download.openmmlab.com/mmcv/dist/cu110/torch1.7.1/index.html
cd CSFwinformer
pip install -e . pip install -r requirements/optional.txt
mkdir data
```
Data Preparation
You can download zip files for corresponding three datasets from "here"
Train
python tools/train.py configs/mirror/pmd_mirror_swin_small.py
Test
python ./tools/test.py configs/mirror/pmd_mirror_swin_small.py work_dirs/pmd_mirror_swin_small/your_weight --show-dir ./results/pmd --eval mIoU
Results and Models
| Dataset | Backbone| IoU↑ | Acc↑ | $Fβ$↑ | MAE↓ | BER↓ | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | PMD | swins | 69.84 | 77.28 | 0.849 | 0.024 | 11.91 | | PMD | swinb | 70.05 | 78.27 | 0.838 | 0.024 | 11.41 | | MSD | swins | 82.13 | 88.72 | 0.895 | 0.046 | 7.15 | | MSD | swinb | 82.08 | 88.92 | 0.896 | 0.045 | 7.14 | | RGBD-Mirror | swinb | 78.66 | 84.64 | 0.900 | 0.031 | 8.57 |
You can find all weights from "here"
Citation
If you find this repo useful for your research, please consider citing our paper:
@ARTICLE{10462920,
author={Xie, Zhifeng and Wang, Sen and Yu, Qiucheng and Tan, Xin and Xie, Yuan},
journal={IEEE Transactions on Image Processing},
title={CSFwinformer: Cross-Space-Frequency Window Transformer for Mirror Detection},
year={2024},
volume={33},
number={},
pages={1853-1867},
keywords={Mirrors;Feature extraction;Transformers;Frequency-domain analysis;Visualization;Semantics;Image segmentation;Mirror detection;texture analysis;cross-modality learning;frequency learning},
doi={10.1109/TIP.2024.3372468}}
Owner
- Login: wangsen99
- Kind: user
- Location: Earth
- Repositories: 2
- Profile: https://github.com/wangsen99
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." authors: - name: "MMSegmentation Contributors" title: "OpenMMLab Semantic Segmentation Toolbox and Benchmark" date-released: 2020-07-10 url: "https://github.com/open-mmlab/mmsegmentation" license: Apache-2.0
GitHub Events
Total
- Issues event: 2
- Watch event: 1
- Issue comment event: 2
Last Year
- Issues event: 2
- Watch event: 1
- Issue comment event: 2