Skip to content

dzp2095/VISTA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

VISTA: Multimodal Test-Time Adaptation

This work has been early accepted by MICCAI 2026. The paper and BibTeX entry will be updated once the arXiv preprint or official proceedings version is available.

Datasets: BraTS (BraTS25-GLIPRE as source; BraTS23-SSA and BraTS24-PED as target). Set csv_path and save_dir in configs/dataset/*.yaml and configs/task/*.yaml to your data and output paths.

Setup: pip install -r requirements.txt

Train source model (e.g. BraTS):
python main.py task=brats_source dataset=brats model=unet training=default

Run TTA (example):
python main.py task=brats_ours dataset=brats_SSA model=unet tta.ckpt_path=/path/to/best_model.pth training.epochs=1 training.batch_size=1

See train_brats.sh.

Citation

If you find this repository useful, please consider citing our paper. The BibTeX entry will be added once the arXiv preprint or official MICCAI proceedings version is available.

% Coming soon.

About

Official code for VISTA.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors