Movie101

About

Movie101 is a large-scale benchmark providing video-aligned movie narration texts to facilitate research on AI movie understanding, like narration generation and temporal grounding. Movie101v2 builds upon Movie101 with bilingual narrations, increased scale, and improved data quality.

Find more details in our papers:

Movie101v2: Improved Movie Narration Benchmark 🔥
Movie101: A New Movie Understanding Benchmark (ACL 2023)

To access the Movie101/Movie101v2 data:

• Annotation: available at the corresponding folders.
• Video: visit our Movie101 homepage.

This repository also contains the evaluation scripts and baseline codes.

Citation

If you find Movie101 useful, please consider citing our papers:

@inproceedings{yue-etal-2023-movie101,
    title = "Movie101: A New Movie Understanding Benchmark",
    author={Zihao Yue and Qi Zhang and Anwen Hu and Liang Zhang and Ziheng Wang and Qin Jin},
    booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    year = "2023",
    url = "https://aclanthology.org/2023.acl-long.257",
    doi = "10.18653/v1/2023.acl-long.257",
    pages = "4669--4684",
}

@misc{yue2024movie101v2,
      title={Movie101v2: Improved Movie Narration Benchmark}, 
      author={Zihao Yue and Yepeng Zhang and Ziheng Wang and Qin Jin},
      year={2024},
      eprint={2404.13370},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}