Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/windygoo/PromptNucSeg
Code for paper "Unleashing the Power of Prompt-driven Nucleus Instance Segmentation"
https://github.com/windygoo/PromptNucSeg
Last synced: about 1 month ago
JSON representation
Code for paper "Unleashing the Power of Prompt-driven Nucleus Instance Segmentation"
- Host: GitHub
- URL: https://github.com/windygoo/PromptNucSeg
- Owner: windygoo
- Created: 2023-11-14T11:18:27.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2024-02-02T07:34:32.000Z (11 months ago)
- Last Synced: 2024-02-02T08:46:18.397Z (11 months ago)
- Language: Python
- Homepage:
- Size: 1.2 MB
- Stars: 24
- Watchers: 4
- Forks: 2
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- Awesome-Segment-Anything - [code
README
## Requirements
torch 2.0.1, mmcv, mmdet
## Dataset
1. download [Kumar](https://github.com/honglianghe/CDNet/issues/6), [CPM-17](https://drive.google.com/drive/folders/1sJ4nmkif6j4s2FOGj8j6i_Ye7z9w0TfA?usp=drive_link) and [PanNuke](https://warwick.ac.uk/fac/cross_fac/tia/data/pannuke) datasets.
2. run **extract_data.py** for data pre-processing.
```markdown
datasets
├── cpm17
│ ├── extract_data.py
│ ├── test
│ └── train
├── cpm17_test_files.npy
├── cpm17_train_files.npy
├── kumar
│ ├── extract_data.py
│ ├── images
│ └── labels
├── kumar_test_files.npy
├── kumar_train_files.npy
├── pannuke
│ ├── extract_data.py
│ ├── Fold 1
│ ├── Fold 2
│ ├── Fold 3
│ ├── Images
│ └── Masks
├── pannuke123_test_files.npy
├── pannuke123_train_files.npy
├── pannuke123_val_files.npy
├── pannuke213_test_files.npy
├── pannuke213_train_files.npy
├── pannuke213_val_files.npy
├── pannuke321_test_files.npy
├── pannuke321_train_files.npy
└── pannuke321_val_files.npy
```
## Training
1. Train the prompter
```shell
cd prompter
python main.py --config dpa_pannuke123.py --output_dir dpa_pannuke123
# python main.py --config dpa_pannuke213.py --output_dir dpa_pannuke123
# python main.py --config dpa_pannuke321.py --output_dir dpa_pannuke123
```
2. Use the trained prompter to generate nuclei prompts for the validation and test sets.
```shell
python predict_prompts.py --config dpa_pannuke123.py --resume checkpoint/dpa_pannuke123/best.pth
# python predict_prompts.py --config dpa_pannuke213.py --resume checkpoint/dpa_pannuke213/best.pth
# python predict_prompts.py --config dpa_pannuke321.py --resume checkpoint/dpa_pannuke321/best.pth
```
3. Download SAM's pre-trained [weights](https://github.com/facebookresearch/segment-anything) into **segmentor/pretrained** and train the segmentor.
```shell
cd segmentor
torchrun --nproc_per_node=4 main.py --config pannuke123_b.py --output_dir pannuke123_b
# torchrun --nproc_per_node=4 main.py --config pannuke213_b.py --output_dir pannuke213_b
# torchrun --nproc_per_node=4 main.py --config pannuke321_b.py --output_dir pannuke321_b
```
## Evaluation
see [test.sh](https://github.com/windygoo/PromptNucSeg/blob/main/segmentor/test.sh)
## Checkpoints
| | Kumar | CPM-17 | PanNuke123 | PanNuke213 | PanNuke321 |
| :---------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
| Prompter | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/Ee_5mPeYZIhGufpsumWbp1QBWPKLg6BxLoXoOzl9BGywVw?e=mHm8Wg) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/Ec0xXaiuz2JIjDInHq1tuEwBJKowhkaxUEqPUiQENeHmPA?e=DSynNh) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EYtVl95nSypFvTJa8B5vSUIB9ibmgxwF9ACFNnDdjBWDXA?e=EvH5PS) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EbmGEmoL539HkBBPpC-SyagB4niZG9IlaNnF71mRuFqa7Q?e=9jHWY5) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EXskYXQgtFZOtu2t-FLzbC8BTqZr8QtRiqtcmOiWCZcpNg?e=05qYTo) |
| Segmentor-B | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EdDo45KYM9BPl3JGKsYaOZsB5lZOUzCZdy7jwZBxn4htGg?e=35cLRu) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EcgysPKrQP1Fs_oByRWAEngBuDjw3Kn6akZbtTl6Wj2hYg?e=nAb6za) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EfznT2AbW5VNiQIIfeq5h8sBvdoioMH35P9PA7bnF1igCQ) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EVvfF959JptOolf7xt8ZPdoBIGIM9UwpTCWDhSLVTtDV_w?e=bJuHZ5) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EQi4RBhTdvFGqIgNyj9UZ4QBTrmK9kJcLwJ4HMlPTHq53w?e=11r7IN) |
| Segmentor-L | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EYLBmedg0nlAp5dqitn8pxcBo_9OcRWHOKpzb5Q9g5f8Kw?e=kie9IK) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/ET0D1YyinExLnum2L3y4soABiPgw_99AcocruqM4bw95pA?e=XVoDhq) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EWQ_o9jIIWVItezvJnpPmkEBQY38Agh0YGHlOHCQZGAIig?e=Foscbm) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EZ02oBK828dLo5P1Z1N9RV0BpzIum-8du7HXDCU4Ue8omg?e=Kt5v5r) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EWErh4qZWSxErgGGxZ_fQPQB3KXnGZ1iTJVtzwwn5sNJyg?e=8ZQo9m) |
| Segmentor-H | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EYCdndKjn4NJv3Qvebo4YsQBrUhU_Uu2tjtBucJH2SMdNQ?e=NThF4d) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/Ebg9v0HaOFZIpyda-JKNST8B2AmnGdhgYQqjdLHYm4j5LA?e=ibANRv) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EZjmiotww1hHtF83WwJTVz0BAmfDNkuSuGbUXkthP3yvDQ?e=N45aU3) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EQG3IMH1OARPj67mapQoakYBjlkMzAKzQjYxPn425JiVeQ?e=6XrKmT) | [OneDrive](https://westlakeu-my.sharepoint.com/:u:/g/personal/shuizhongyi_westlake_edu_cn/EVj-vCQh5MVPqIT8ggkSJGsBeWv_MsrO9Ci3Lr7wuewW2A?e=qsa0Gd) |
## Contact
If you have any questions or concerns, feel free to report issues or directly contact us (Zhongyi Shui [email protected]).
## Citation
If you find this code useful for your research, please cite us using the following BibTeX entry.
```bibtex
@article{shui2023unleashing,
title={Unleashing the Power of Prompt-driven Nucleus Instance Segmentation},
author={Shui, Zhongyi and Zhang, Yunlong and Yao, Kai and Zhu, Chenglu and Sun, Yuxuan and Yang, Lin},
journal={arXiv preprint arXiv:2311.15939},
year={2023}
}
```