Technical Report for Soccernet 2023 -- Dense Video Captioning
Zheng Ruan, Ruixuan Liu, Shimin Chen, Mengying Zhou, Xinquan Yang, Wei Li, Chen Chen, Wei Shen
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.
Related Papers
UGC-VideoCaptioner: An Omni UGC Video Detail Caption Model and New Benchmarks2025-07-15Show, Tell and Summarize: Dense Video Captioning Using Visual Cue Aided Sentence Summarization2025-06-25Dense Video Captioning using Graph-based Sentence Summarization2025-06-25video-SALMONN 2: Captioning-Enhanced Audio-Visual Large Language Models2025-06-18VersaVid-R1: A Versatile Video Understanding and Reasoning Model from Question Answering to Captioning Tasks2025-06-10ARGUS: Hallucination and Omission Evaluation in Video-LLMs2025-06-09Temporal Object Captioning for Street Scene Videos from LiDAR Tracks2025-05-22FLASH: Latent-Aware Semi-Autoregressive Speculative Decoding for Multimodal Tasks2025-05-19