CVPR 2025
- Download XACV from google drive
- Replace
xca_datasetwith our XACV dataset.
XACV
├── CVAI-2828
│ ├── ground_truth
│ │ ├── CVAI-2828RAO2_CRA32
│ │ │ ├── 00056.png
│ │ │ └── 00065.png
│ │ └── CVAI-2828RAO2_CRA32CATH
│ │ ├── 00056.png
│ │ └── 00065.png
│ └── images
│ └── CVAI-2828RAO2_CRA32
│ ├── 00000.png
│ ├── 00001.png
│ ├── 00002.png
│ ├── 00003.png
│ ├── 00004.png
│ ├── ...
├── CVAI-2829
├── ...
Python versionn = 3.9
conda env create -f environment.yml
cd RAFT
./download_models.sh
python main.py -d {seq_name}
e.g.
python main.py -d CVAI-2828RAO2_CRA32
python launch.py -f ./job_specs/vessel.txt --gpus {gpu_ids}
cd eval
python gt_filename.py -d {exp_name}
python eval.py
exp_name is the last directory name in the dir path specified in confs/config.yaml.
e.g.
python gt_filename.py -d init_model
Please cite us if our work is useful for your research.
@inproceedings{wu2025denver,
title={DeNVeR: Deformable Neural Vessel Representations for Unsupervised Video Vessel Segmentation},
author={Wu, Chun-Hung and Chen, Shih-Hong and Hu, Chih Yao and Wu, Hsin-Yu and Chen, Kai-Hsin and Chen, Yu-You and Su, Chih-Hai and Lee, Chih-Kuo and Liu, Yu-Lun},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year={2025}
}
This research was funded by the National Science and Technology Council, Taiwan, under Grants NSTC 112-2222-E-A49-004-MY2. The authors are grateful to Google, NVIDIA, and MediaTek Inc. for generous donations. Yu-Lun Liu acknowledges the Yushan Young Fellow Program by the MOE in Taiwan.