Image-guided topic modeling for interpretable privacy classification

TL;DR

This paper introduces a novel image-guided topic modeling approach that generates natural language descriptors to predict and explain image privacy, enhancing interpretability and performance over existing methods.

Contribution

The paper presents a new multimodal image-guided topic modeling technique (ITM) that produces interpretable descriptors for privacy prediction, outperforming previous interpretable models.

Findings

Priv×ITM classifier outperforms existing interpretable methods by 5% in accuracy.

Priv×ITM performs comparably to non-interpretable state-of-the-art models.

ITM effectively leverages vision and language data for privacy classification.

Abstract

Predicting and explaining the private information contained in an image in human-understandable terms is a complex and contextual task. This task is challenging even for large language models. To facilitate the understanding of privacy decisions, we propose to predict image privacy based on a set of natural language content descriptors. These content descriptors are associated with privacy scores that reflect how people perceive image content. We generate descriptors with our novel Image-guided Topic Modeling (ITM) approach. ITM leverages, via multimodality alignment, both vision information and image textual descriptions from a vision language model. We use the ITM-generated descriptors to learn a privacy predictor, Priv$\times$ITM, whose decisions are interpretable by design. Our Priv$\times$ITM classifier outperforms the reference interpretable method by 5 percentage points in accuracy and performs comparably to the current non-interpretable state-of-the-art model.