Published October 29, 2024 | Version v1
Dataset Open

EventCap

Authors/Creators

  • 1. ROR icon Ho Chi Minh City University of Science

Description

 
The dataset EventCap, includes 3,140 image-caption pairs centered around real-world events. Each pair has been meticulously curated from images and information drawn from a diverse selection of reputable articles. This dataset provides a robust resource for training and assessing the capabilities of image captioning models in accurately interpreting and describing complex events in real-world contexts.

 

Notes

If you use our dataset, code, or any part of the VisChronos framework in your work, please cite the following paper:

@inproceedings{nguyen2024vischronos,
  title={VisChronos: Revolutionizing Image Captioning Through Real-Life Events},
  author={Nguyen, Phuc-Tan and Nguyen, Hieu and Le, Trung-Nghia},
  booktitle={International Symposium on Information and Communication Technology},
  pages={127--140},
  year={2024},
  organization={Springer}
}

Files

label.csv

Files (1.8 GB)

Name Size Download all
md5:34be0b83296af526a2434e4ab9dc7ef7
6.9 MB Preview Download
md5:66aec24ef3405c8b09ba63672296cc8e
1.8 GB Preview Download

Additional details

References

  • Phuc-Tan Nguyen, Hieu Nguyen, Minh-Triet Tran, Trung-Nghia Le, "VisChronos: Revolutionizing Image Captioning Through Real-Life Events", SOICT, 2024.