Video Joint Modelling Based on Hierarchical Transformer for Co-summarization

TL;DR

This paper introduces VJMHT, a hierarchical transformer model that jointly models multiple videos to improve summarization by capturing cross-video semantic dependencies, leading to more informative summaries.

Contribution

The paper proposes a novel hierarchical transformer framework for co-summarization that explicitly models semantic dependencies across similar videos.

Findings

VJMHT outperforms existing methods in F-measure and ranking evaluations.

Cross-video semantic modeling improves summarization quality.

Transformer-based representation reconstruction enhances summary fidelity.

Abstract

Video summarization aims to automatically generate a summary (storyboard or video skim) of a video, which can facilitate large-scale video retrieval and browsing. Most of the existing methods perform video summarization on individual videos, which neglects the correlations among similar videos. Such correlations, however, are also informative for video understanding and video summarization. To address this limitation, we propose Video Joint Modelling based on Hierarchical Transformer (VJMHT) for co-summarization, which takes into consideration the semantic dependencies across videos. Specifically, VJMHT consists of two layers of Transformer: the first layer extracts semantic representation from individual shots of similar videos, while the second layer performs shot-level video joint modelling to aggregate cross-video semantic information. By this means, complete cross-video high-level patterns are explicitly modelled and learned for the summarization of individual videos. Moreover, Transformer-based video representation reconstruction is introduced to maximize the high-level similarity between the summary and the original video. Extensive experiments are conducted to verify the effectiveness of the proposed modules and the superiority of VJMHT in terms of F-measure and rank-based evaluation.