MC-PanDA: Mask Confidence for Panoptic Domain Adaptation (ECCV 2024)

This repo contains the code for our paper MC-PanDA: Mask Confidence for Panoptic Domain Adaptation.

Installation

See installation instructions for MC-PanDA.

Getting Started

• See Datasets preparation for MC-PanDA.
• See training and eval scripts for MC-PanDA.

Model Zoo

See MC-PanDA MODEL ZOO.

Acknowledgments

This project is based on the following open-source projects. We thank their authors for making the source code publically available:

• Mask2Former
• detectron2
• EDAPS

Citing MC-PanDA

If you use MC-PanDA in your research, please use the following BibTeX entry.

@inproceedings{martinovic2024eccv,
  title={MC-PanDA: Mask Confidence for Panoptic Domain Adaptation},
  author={Martinovi{\'c}, Ivan and {\v{S}}ari{\'c}, Josip and {\v{S}}egvi{\'c}, Sini{\v{s}}a},
  booktitle={European Conference on Computer Vision},
  pages={167--185},
  year={2025},
  organization={Springer}
}