Telling the What while Pointing to the Where: Multimodal Queries for Image Retrieval

TL;DR

This paper introduces a multimodal image retrieval approach combining spoken language and mouse traces to better specify content and location, significantly improving retrieval accuracy over text-only systems.

Contribution

It proposes a novel multimodal query system using speech and pointing for image retrieval and adapts existing models to incorporate spatial guidance.

Findings

Model effectively uses spatial guidance for improved accuracy.

Significant performance gains over text-only retrieval systems.

Qualitative and quantitative results validate the approach.

Abstract

Most existing image retrieval systems use text queries as a way for the user to express what they are looking for. However, fine-grained image retrieval often requires the ability to also express where in the image the content they are looking for is. The text modality can only cumbersomely express such localization preferences, whereas pointing is a more natural fit. In this paper, we propose an image retrieval setup with a new form of multimodal queries, where the user simultaneously uses both spoken natural language (the what) and mouse traces over an empty canvas (the where) to express the characteristics of the desired target image. We then describe simple modifications to an existing image retrieval model, enabling it to operate in this setup. Qualitative and quantitative experiments show that our model effectively takes this spatial guidance into account, and provides significantly more accurate retrieval results compared to text-only equivalent systems.