Dense Video Captioning using Graph-based Sentence Summarization

TL;DR

This paper introduces a graph-based framework for dense video captioning that improves scene understanding by summarizing sequences of video segments into comprehensive descriptions, leveraging semantic word relationships.

Contribution

The paper proposes a novel GPaS framework with a GCN-LSTM interaction module for better scene evolution modeling in dense video captioning, focusing on the summarization stage.

Findings

Outperforms state-of-the-art on ActivityNet Captions dataset

Effective in capturing scene evolution over long proposals

Improves caption quality by exploiting semantic word relationships

Abstract

Recently, dense video captioning has made attractive progress in detecting and captioning all events in a long untrimmed video. Despite promising results were achieved, most existing methods do not sufficiently explore the scene evolution within an event temporal proposal for captioning, and therefore perform less satisfactorily when the scenes and objects change over a relatively long proposal. To address this problem, we propose a graph-based partition-and-summarization (GPaS) framework for dense video captioning within two stages. For the ``partition" stage, a whole event proposal is split into short video segments for captioning at a finer level. For the ``summarization" stage, the generated sentences carrying rich description information for each segment are summarized into one sentence to describe the whole event. We particularly focus on the ``summarization" stage, and propose a framework that effectively exploits the relationship between semantic words for summarization. We achieve this goal by treating semantic words as nodes in a graph and learning their interactions by coupling Graph Convolutional Network (GCN) and Long Short Term Memory (LSTM), with the aid of visual cues. Two schemes of GCN-LSTM Interaction (GLI) modules are proposed for seamless integration of GCN and LSTM. The effectiveness of our approach is demonstrated via an extensive comparison with the state-of-the-arts methods on the two benchmarks ActivityNet Captions dataset and YouCook II dataset.