CoolFace
Datasetpublic

VLM2Vec/DiDeMo

Clone from friedrichor/DiDeMo. About DiDeMo contains 10K long-form videos from Flickr. For each video, ~4 short sentences are annotated in temporal order. We follow the existing works to concatenate those short sentences and evaluate ‘paragraph-to-video’ retrieval on this benchmark. We adopt the official split: Train: 8,395 videos, 8,395 captions (concatenate from 33,005 short captions) Val: 1,065 videos, 1,065 captions (concatenate from 4,290 short captions) (We don't… See the full description on the dataset page: https://huggingface.co/datasets/VLM2Vec/DiDeMo.

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes4.6kdownloads
Dataset Card

Clone from friedrichor/DiDeMo.

About

DiDeMo contains 10K long-form videos from Flickr. For each video, ~4 short sentences are annotated in temporal order. We follow the existing works to concatenate those short sentences and evaluate ‘paragraph-to-video’ retrieval on this benchmark.

We adopt the official split:

  • Train: 8,395 videos, 8,395 captions (concatenate from 33,005 short captions)
  • Val: 1,065 videos, 1,065 captions (concatenate from 4,290 short captions) (We don't have the collection yet.)
  • Test: 1,004 videos, 1,004 captions (concatenate from 4,021 short captions)

Get Raw Videos

bash
cat DiDeMo_Videos_mp4_train.tar.part-* | tar -vxf -
tar -xvf DiDeMo_Videos_mp4_test.tar

Official Release

Video Release: DiDeMoRelease


🌟 Citation

bibtex
@inproceedings{anne2017didemo,
  title={Localizing moments in video with natural language},
  author={Anne Hendricks, Lisa and Wang, Oliver and Shechtman, Eli and Sivic, Josef and Darrell, Trevor and Russell, Bryan},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2017}
}