Cross-Modal Self-Attention Network for Referring Image Segmentation

We consider the problem of referring image segmentation. Given an input image and a natural language expression, the goal is to segment the object referred by the language expression in the image. Existing works in this area treat the language expression and the input image separately in their representations. They do not sufficiently capture long-range correlations between these two modalities. In this paper, we propose a cross-modal self-attention (CMSA) module that effectively captures the long-range dependencies between linguistic and visual features. Our model can adaptively focus on informative words in the referring expression and important regions in the input image. In addition, we propose a gated multi-level fusion module to selectively integrate self-attentive cross-modal features corresponding to different levels in the image. This module controls the information flow of features at different levels. We validate the proposed approach on four evaluation datasets. Our proposed approach consistently outperforms existing state-of-the-art methods.

1. Introduction

Referring image segmentation is a challenging problem at the intersection of computer vision and natural language processing. Given an image and a natural language expression, the goal is to produce a segmentation mask in the image corresponding to entities referred by the the natural language expression (see Fig. 4 for some examples). It is worth noting that the referring expression is not limited to specifying object categories (e.g. "person", "cat"). It can take any free form language description which may contain appearance attributes (e.g. "red", "long"), actions (e.g. "standing", "hold") and relative relationships (e.g. "left", "above"), etc. Referring image segmentation can potentially be used in a wide range of applications, such as interactive photo editing and human-robot interaction.