MAttNet: Modular Attention Network for Referring Expression Comprehension

TL;DR

This paper introduces MAttNet, a modular attention network that decomposes referring expressions into subject, location, and relationship components, enabling flexible and improved image region localization.

Contribution

The paper presents a novel end-to-end modular attention framework that dynamically combines components for better referring expression comprehension.

Findings

Outperforms previous state-of-the-art methods significantly

Effective decomposition of expressions improves localization accuracy

Demonstrates flexibility in handling diverse expression types

Abstract

In this paper, we address referring expression comprehension: localizing an image region described by a natural language expression. While most recent work treats expressions as a single unit, we propose to decompose them into three modular components related to subject appearance, location, and relationship to other objects. This allows us to flexibly adapt to expressions containing different types of information in an end-to-end framework. In our model, which we call the Modular Attention Network (MAttNet), two types of attention are utilized: language-based attention that learns the module weights as well as the word/phrase attention that each module should focus on; and visual attention that allows the subject and relationship modules to focus on relevant image components. Module weights combine scores from all three modules dynamically to output an overall score. Experiments show that MAttNet outperforms previous state-of-art methods by a large margin on both bounding-box-level and pixel-level comprehension tasks. Demo and code are provided.