UBiSS: A Unified Framework for Bimodal Semantic Summarization of Videos

TL;DR

UBiSS introduces a unified framework for bimodal video summarization that generates both visual and textual summaries simultaneously, leveraging a large-scale dataset and a novel evaluation metric to improve semantic content preservation.

Contribution

The paper presents a new large-scale dataset BIDS and a unified model UBiSS for bimodal semantic video summarization, advancing beyond traditional unimodal and multi-stage methods.

Findings

UBiSS outperforms multi-stage pipelines in summarization quality.

The BIDS dataset effectively captures salient content in long videos.

The proposed NDCG_MS metric provides comprehensive evaluation of bimodal summaries.

Abstract

With the surge in the amount of video data, video summarization techniques, including visual-modal(VM) and textual-modal(TM) summarization, are attracting more and more attention. However, unimodal summarization inevitably loses the rich semantics of the video. In this paper, we focus on a more comprehensive video summarization task named Bimodal Semantic Summarization of Videos (BiSSV). Specifically, we first construct a large-scale dataset, BIDS, in (video, VM-Summary, TM-Summary) triplet format. Unlike traditional processing methods, our construction procedure contains a VM-Summary extraction algorithm aiming to preserve the most salient content within long videos. Based on BIDS, we propose a Unified framework UBiSS for the BiSSV task, which models the saliency information in the video and generates a TM-summary and VM-summary simultaneously. We further optimize our model with a list-wise ranking-based objective to improve its capacity to capture highlights. Lastly, we propose a metric, $NDCG_{MS}$, to provide a joint evaluation of the bimodal summary. Experiments show that our unified framework achieves better performance than multi-stage summarization pipelines. Code and data are available at https://github.com/MeiYutingg/UBiSS.