MS-DPPs: Multi-Source Determinantal Point Processes for Contextual Diversity Refinement of Composite Attributes in Text to Image Retrieval

TL;DR

This paper introduces MS-DPPs, a novel multi-source determinantal point process model that refines attribute diversity in text-to-image retrieval based on application context, improving result diversification.

Contribution

It extends DPPs to handle multiple sources with a unified similarity matrix and introduces Tangent Normalization for contextual diversity refinement.

Findings

MS-DPPs outperform existing methods in diversity refinement tasks.

The approach effectively models multi-source attribute diversity.

Experimental results validate the method's superiority in context-aware diversification.

Abstract

Result diversification (RD) is a crucial technique in Text-to-Image Retrieval for enhancing the efficiency of a practical application. Conventional methods focus solely on increasing the diversity metric of image appearances. However, the diversity metric and its desired value vary depending on the application, which limits the applications of RD. This paper proposes a novel task called CDR-CA (Contextual Diversity Refinement of Composite Attributes). CDR-CA aims to refine the diversities of multiple attributes, according to the application's context. To address this task, we propose Multi-Source DPPs, a simple yet strong baseline that extends the Determinantal Point Process (DPP) to multi-sources. We model MS-DPP as a single DPP model with a unified similarity matrix based on a manifold representation. We also introduce Tangent Normalization to reflect contexts. Extensive experiments demonstrate the effectiveness of the proposed method. Our code is publicly available at https://github.com/NEC-N-SOGI/msdpp.