JanySunny / CRVOS

Official pytorch code for "CRVOS: Clue Refining Network for Video Object Segmentation"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

CRVOS: Clue Refining Network for Video Object Segmentation

PWC

Suhwan Cho, MyeongAh Cho, Tae-young Chung, Heansung Lee, Sangyoun Lee

arXiv paper: [pdf]


How to run

  1. Download DAVIS 2017 and pre-trained model.

  2. Modify "local_config.py".

  3. Check fps of CRVOS.

python test.py --fps
  1. Save the outputs of CRVOS.
python test.py --save

How to evaluate

Official evaluation code can only deal with DAVIS 2017.

If you want to evaluate the model on both DAVIS 2016 and DAVIS 2017, use my code.

[Official evaluation code]

[My evaluation code]

Citation

@article{cho2020crvos,
  title={CRVOS: Clue Refining Network for Video Object Segmentation},
  author={Cho, Suhwan and Cho, MyeongAh and Chung, Tae-young and Lee, Heansung and Lee, Sangyoun},
  journal={arXiv preprint arXiv:2002.03651},
  year={2020}
}

Acknowledgements

This code is built on joakimjohnander/agame-vos. We appreciate the authors sharing the code.

About

Official pytorch code for "CRVOS: Clue Refining Network for Video Object Segmentation"


Languages

Language:Python 100.0%