English

Technical Report for Soccernet 2023 -- Dense Video Captioning

Computer Vision and Pattern Recognition 2024-11-05 v1

Abstract

In the task of dense video captioning of Soccernet dataset, we propose to generate a video caption of each soccer action and locate the timestamp of the caption. Firstly, we apply Blip as our video caption framework to generate video captions. Then we locate the timestamp by using (1) multi-size sliding windows (2) temporal proposal generation and (3) proposal classification.

Cite

@article{arxiv.2411.00882,
  title  = {Technical Report for Soccernet 2023 -- Dense Video Captioning},
  author = {Zheng Ruan and Ruixuan Liu and Shimin Chen and Mengying Zhou and Xinquan Yang and Wei Li and Chen Chen and Wei Shen},
  journal= {arXiv preprint arXiv:2411.00882},
  year   = {2024}
}
R2 v1 2026-06-28T19:44:46.675Z