DatasetHarvester

A collection of scripts to easily download Video Object Segmentation (VOS) datasets for my research.

---

Note

Contributions are welcomed to improve this repo 🤗

---

- Download the SA-V Dataset

Introduced in SAM 2: Segment Anything in Images and Videos Paper / Code

1. Fill this Meta document out to get access to the URLs
2. Adapt the corresponding download_SA_V.yaml with the URLs
3. Adapt the paths in download_SA_V.py - line 44
4. Run

   cd scripts/SA_V/
   python download_SA_V.py

5. More stuff HERE

• Citation

  @article{ravi2024sam2,
    title={SAM 2: Segment Anything in Images and Videos},
    author={Ravi, Nikhila and Gabeur, Valentin and Hu, Yuan-Ting and Hu, Ronghang and Ryali, Chaitanya and Ma, Tengyu and Khedr, Haitham and R{\"a}dle, Roman and Rolland, Chloe and Gustafson, Laura and Mintun, Eric and Pan, Junting and Alwala, Kalyan Vasudev and Carion, Nicolas and Wu, Chao-Yuan and Girshick, Ross and Doll{\'a}r, Piotr and Feichtenhofer, Christoph},
    journal={arXiv preprint arXiv:2408.00714},
    url={https://arxiv.org/abs/2408.00714},
    year={2024}
  }

Note

It took me circa 3h30min to download the complete dataset (using a single core)

- Download the LVOS Dataset (v1 and v2)

Introduced in LVOS: A Benchmark for Large-scale Long-term Video Object Segmentation

1. Adapt the path in l. 15
2. Run

   cd scripts/LVOS/
   python download_LVOS.py

   Optional: Run the following script if needed

   python annotation_first_only.py

3. Official Evaluation ToolKit

• Citation

  @InProceedings{Hong_2023_ICCV,
      author    = {Hong, Lingyi and Chen, Wenchao and Liu, Zhongying and Zhang, Wei and Guo, Pinxue and Chen, Zhaoyu and Zhang, Wenqiang},
      title     = {LVOS: A Benchmark for Long-term Video Object Segmentation},
      booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
      month     = {October},
      year      = {2023},
      pages     = {13480-13492}
  }

Note

Don't run too many times on the same day the download on the same PC...

- Download the VOST Dataset