Commonsense Knowledge Aware Concept Selection For Diverse and Informative Visual Storytelling

TL;DR

This paper introduces a concept selection approach leveraging a commonsense knowledge graph to enhance the diversity and informativeness of visual storytelling, outperforming previous models in relevance and quality.

Contribution

It proposes a novel concept selection module using a commonsense knowledge graph and correlation modules to improve story diversity and informativeness in visual storytelling.

Findings

Outperforms previous models in story diversity and informativeness

Produces more relevant and reasonable concepts for storytelling

Enhances visual storytelling quality through knowledge graph integration

Abstract

Visual storytelling is a task of generating relevant and interesting stories for given image sequences. In this work we aim at increasing the diversity of the generated stories while preserving the informative content from the images. We propose to foster the diversity and informativeness of a generated story by using a concept selection module that suggests a set of concept candidates. Then, we utilize a large scale pre-trained model to convert concepts and images into full stories. To enrich the candidate concepts, a commonsense knowledge graph is created for each image sequence from which the concept candidates are proposed. To obtain appropriate concepts from the graph, we propose two novel modules that consider the correlation among candidate concepts and the image-concept correlation. Extensive automatic and human evaluation results demonstrate that our model can produce reasonable concepts. This enables our model to outperform the previous models by a large margin on the diversity and informativeness of the story, while retaining the relevance of the story to the image sequence.