In this repository, we present the code of SAM2-based Mask Enhancement and Conditional Mask Fusion, which corresponds to the 2nd and 3rd stages of our solution.
Meanwhile, we provide our intermediate results from the 1st stage on Google Drive, which can be directly used to execute this code.
The code of ReferDINO will be available at here once it is ready, thanks for your patience 🫡.
Download the pretrained SAM 2 checkpoints:
cd checkpoints
bash download_ckpts.shor individually from:
The code requires python>=3.10, as well as torch>=2.5.1 and torchvision>=0.20.1. Please follow the instructions here to install both PyTorch and TorchVision dependencies. You can install SAM 2 on a GPU machine using:
cd sam2
pip install -e .
pip install -r requirements.txtExecuting the code is straightforward:
python refine_from_refdino.py --gids 0 1 2 3 4 5 6 7The parameter gids should be set based on the number of GPUs available on your device. The code will automatically create a refine_output directory and save the results under it.
Our code is based on SAM2. It is a fantastic work.
If you find our work helpful for your research, please consider citing our report and paper.
@inproceedings{liang2025referdino,
title={ReferDINO: Referring Video Object Segmentation with Visual Grounding Foundations},
author={Liang, Tianming and Lin, Kun-Yu and Tan, Chaolei and Zhang, Jianguo and Zheng, Wei-Shi and Hu, Jian-Fang},
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
year={2025}
}
@article{liang2025referdinoplus,
title={ReferDINO-Plus: 2nd Solution for 4th PVUW MeViS Challenge at CVPR 2025},
author={Liang, Tianming and Jiang, Haichao and Zheng, Wei-Shi and Hu, Jian-Fang},
journal={arXiv preprint arXiv:2503.23509},
year={2025}
}