Home

Awesome

MOSE: A New Dataset for Video Object Segmentation in Complex Scenes

šŸ [Homepage] ā€ƒ šŸ“„[Arxiv]

This repository contains information and tools for the MOSE dataset.

Download

[šŸ”„02.09.2023: Dataset has been released!]

ā¬‡ļø Get the dataset from:

šŸ“¦ Or use gdown:

# train.tar.gz
gdown 'https://drive.google.com/uc?id=ID_removed_to_avoid_overaccesses_get_it_by_yourself'

# valid.tar.gz
gdown 'https://drive.google.com/uc?id=ID_removed_to_avoid_overaccesses_get_it_by_yourself'

# test set will be released when competition starts.

Please also check the SHA256 sum of the files to ensure the data intergrity:

3f805e66ecb576fdd37a1ab2b06b08a428edd71994920443f70d09537918270b train.tar.gz
884baecf7d7e85cd35486e45d6c474dc34352a227ac75c49f6d5e4afb61b331c valid.tar.gz

Evaluation

[šŸ”„02.16.2023: Our CodaLab competition is on live now!]

Please submit your results on

File Structure

The dataset follows a similar structure as DAVIS and Youtube-VOS. The dataset consists of two parts: JPEGImages which holds the frame images, and Annotations which contains the corresponding segmentation masks. The frame images are numbered using five-digit numbers. Annotations are saved in color-pattlate mode PNGs like DAVIS.

Please note that while annotations for all frames in the training set are provided, annotations for the validation set will only include the first frame.

<train/valid.tar>
ā”‚
ā”œā”€ā”€ Annotations
ā”‚ ā”‚ 
ā”‚ ā”œā”€ā”€ <video_name_1>
ā”‚ ā”‚ ā”œā”€ā”€ 00000.png
ā”‚ ā”‚ ā”œā”€ā”€ 00001.png
ā”‚ ā”‚ ā””ā”€ā”€ ...
ā”‚ ā”‚ 
ā”‚ ā”œā”€ā”€ <video_name_2>
ā”‚ ā”‚ ā”œā”€ā”€ 00000.png
ā”‚ ā”‚ ā”œā”€ā”€ 00001.png
ā”‚ ā”‚ ā””ā”€ā”€ ...
ā”‚ ā”‚ 
ā”‚ ā”œā”€ā”€ <video_name_...>
ā”‚ 
ā””ā”€ā”€ JPEGImages
  ā”‚ 
  ā”œā”€ā”€ <video_name_1>
  ā”‚ ā”œā”€ā”€ 00000.jpg
  ā”‚ ā”œā”€ā”€ 00001.jpg
  ā”‚ ā””ā”€ā”€ ...
  ā”‚ 
  ā”œā”€ā”€ <video_name_2>
  ā”‚ ā”œā”€ā”€ 00000.jpg
  ā”‚ ā”œā”€ā”€ 00001.jpg
  ā”‚ ā””ā”€ā”€ ...
  ā”‚ 
  ā””ā”€ā”€ <video_name_...>

BibTeX

Please consider to cite MOSE if it helps your research.

@inproceedings{MOSE,
  title={{MOSE}: A New Dataset for Video Object Segmentation in Complex Scenes},
  author={Ding, Henghui and Liu, Chang and He, Shuting and Jiang, Xudong and Torr, Philip HS and Bai, Song},
  booktitle={ICCV},
  year={2023}
}

License

MOSE is licensed under a CC BY-NC-SA 4.0 License. The data of MOSE is released for non-commercial research purpose only.