Whether you can locate or not? Interactive Referring Expression Generation

TL;DR

This paper introduces an Interactive Referring Expression Generation (IREG) model that collaborates with a Referring Expression Comprehension (REC) model, improving RE quality through interaction and outperforming previous methods on benchmark datasets.

Contribution

The paper presents the first interactive REG model that leverages REC signals to iteratively refine referring expressions, enhancing accuracy and human preference.

Findings

IREG outperforms state-of-the-art methods on benchmark datasets

Interaction with REC improves RE quality and relevance

Human evaluation favors IREG-generated expressions

Abstract

Referring Expression Generation (REG) aims to generate unambiguous Referring Expressions (REs) for objects in a visual scene, with a dual task of Referring Expression Comprehension (REC) to locate the referred object. Existing methods construct REG models independently by using only the REs as ground truth for model training, without considering the potential interaction between REG and REC models. In this paper, we propose an Interactive REG (IREG) model that can interact with a real REC model, utilizing signals indicating whether the object is located and the visual region located by the REC model to gradually modify REs. Our experimental results on three RE benchmark datasets, RefCOCO, RefCOCO+, and RefCOCOg show that IREG outperforms previous state-of-the-art methods on popular evaluation metrics. Furthermore, a human evaluation shows that IREG generates better REs with the capability of interaction.