Searching for Ambiguous Objects in Videos using Relational Referring Expressions

TL;DR

This paper introduces a new dataset and deep learning methods for searching ambiguous objects in videos using relational referring expressions, emphasizing the importance of relational context in disambiguation.

Contribution

The paper presents a novel dataset for video object search with relational expressions and a deep attention network that outperforms baselines in highly ambiguous settings.

Findings

Deep networks trained on the new dataset show promising results.

The proposed attention network significantly outperforms baseline models.

The dataset highlights challenges of non-relational expressions in ambiguous scenarios.

Abstract

Humans frequently use referring (identifying) expressions to refer to objects. Especially in ambiguous settings, humans prefer expressions (called relational referring expressions) that describe an object with respect to a distinguishing, unique object. Unlike studies on video object search using referring expressions, in this paper, our focus is on (i) relational referring expressions in highly ambiguous settings, and (ii) methods that can both generate and comprehend a referring expression. For this goal, we first introduce a new dataset for video object search with referring expressions that includes numerous copies of the objects, making it difficult to use non-relational expressions. Moreover, we train two baseline deep networks on this dataset, which show promising results. Finally, we propose a deep attention network that significantly outperforms the baselines on our dataset. The dataset and the codes are available at https://github.com/hazananayurt/viref.