R$^3$Net:Relation-embedded Representation Reconstruction Network for Change Captioning

TL;DR

This paper introduces R$^3$Net, a novel network that effectively distinguishes real image changes from distractors like viewpoint variations, improving change captioning accuracy through relation-embedded modules and semantic interaction enhancements.

Contribution

The paper presents a relation-embedded module and a representation reconstruction module for better change detection, along with a syntactic skeleton predictor to improve caption generation.

Findings

Achieves state-of-the-art results on two public datasets.

Effectively distinguishes real changes from viewpoint distractors.

Enhances change captioning accuracy with novel modules.

Abstract

Change captioning is to use a natural language sentence to describe the fine-grained disagreement between two similar images. Viewpoint change is the most typical distractor in this task, because it changes the scale and location of the objects and overwhelms the representation of real change. In this paper, we propose a Relation-embedded Representation Reconstruction Network (R$^3$Net) to explicitly distinguish the real change from the large amount of clutter and irrelevant changes. Specifically, a relation-embedded module is first devised to explore potential changed objects in the large amount of clutter. Then, based on the semantic similarities of corresponding locations in the two images, a representation reconstruction module (RRM) is designed to learn the reconstruction representation and further model the difference representation. Besides, we introduce a syntactic skeleton predictor (SSP) to enhance the semantic interaction between change localization and caption generation. Extensive experiments show that the proposed method achieves the state-of-the-art results on two public datasets.