Factorized Multi-Modal Topic Model

TL;DR

This paper introduces a novel HDP-based multi-modal topic model that automatically learns shared and private topics across different data modalities, improving analysis of paired image and text data.

Contribution

It presents a new model that captures both shared and private components in multi-modal data, addressing limitations of previous methods.

Findings

Effectively learns shared and private topics.

Enhances querying across modalities.

Outperforms existing models in multi-modal analysis.

Abstract

Multi-modal data collections, such as corpora of paired images and text snippets, require analysis methods beyond single-view component and topic models. For continuous observations the current dominant approach is based on extensions of canonical correlation analysis, factorizing the variation into components shared by the different modalities and those private to each of them. For count data, multiple variants of topic models attempting to tie the modalities together have been presented. All of these, however, lack the ability to learn components private to one modality, and consequently will try to force dependencies even between minimally correlating modalities. In this work we combine the two approaches by presenting a novel HDP-based topic model that automatically learns both shared and private topics. The model is shown to be especially useful for querying the contents of one domain given samples of the other.