README.md

metadata

configs:
  - config_name: default
    data_files:
      - split: test
        path: m3vos_viewer_data_with_paths.jsonl
license: apache-2.0
task_categories:
  - video-classification
language:
  - en
  - zh
tags:
  - CVPR2025
  - video
  - segmentation
  - computer-vision
  - physical
  - M3-VOS
pretty_name: M3-VOS
size_categories:
  - n<1K

[CVPR 2025] M³-VOS: Multi-Phase, Multi-Transition, and Multi-Scenery Video Object Segmentation

If you like our project, please give us a star ⭐ on GitHub for the latest update.

💡 Description

• Venue: CVPR2025
• Repository: 🛠️Tool, 🏠Page
• Paper: arxiv.org/html/2412.13803v2
• Point of Contact: Jiaxin Li , Zixuan Chen

📁 Structure

This dataset contains annotated videos and images for object segmentation tasks with phase transition information. The directory structure and file descriptions are as follows:

• meta/
  • all_core_seqs.txt: A list of core sequences used in the dataset.
  • all_phase_transition.json: Metadata describing the phase transition states of target objects.
  • target_object.json: Contains information about the target objects in each video sequence.
• data/
  • Annotations/: Contains segmentation masks for the annotated target objects.
  • Videos/: The original video files corresponding to each sequence.
  • JPEGImages/: Extracted image frames from the videos.
  • ImageSets/
    • val.txt: A list of video sequences used for validation.

For more details, please refer to our paper on arXiv: 2412.13803.

✏️ Citation

If you find our paper and code useful in your research, please consider giving a star and citation.

@InProceedings{chen2024m3vos_2025_CVPR,
    author    = {Zixuan Chen and Jiaxin Li and Liming Tan and Yejie Guo and Junxuan Liang and Cewu Lu and Yong-Lu Li},
    title     = {M$^3$-VOS: Multi-Phase, Multi-Transition, and Multi-Scenery Video Object Segmentation},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2025}
}