Video Timeline Modeling For News Story Understanding

TL;DR

This paper introduces the novel task of video timeline modeling for news story understanding, creates a large benchmark dataset, proposes evaluation metrics, and benchmarks deep learning methods to advance research in this area.

Contribution

It defines the new problem of video timeline modeling, provides a comprehensive dataset and metrics, and benchmarks initial deep learning approaches.

Findings

Established a new benchmark dataset with 12k timelines and 300k videos.

Proposed quantitative metrics for evaluating video timeline models.

Benchmarked several deep learning approaches for the task.

Abstract

In this paper, we present a novel problem, namely video timeline modeling. Our objective is to create a video-associated timeline from a set of videos related to a specific topic, thereby facilitating the content and structure understanding of the story being told. This problem has significant potential in various real-world applications, for instance, news story summarization. To bootstrap research in this area, we curate a realistic benchmark dataset, YouTube-News-Timeline, consisting of over $12$k timelines and $300$k YouTube news videos. Additionally, we propose a set of quantitative metrics to comprehensively evaluate and compare methodologies. With such a testbed, we further develop and benchmark several deep learning approaches to tackling this problem. We anticipate that this exploratory work will pave the way for further research in video timeline modeling. The assets are available via https://github.com/google-research/google-research/tree/master/video_timeline_modeling.