Retrieving Black-box Optimal Images from External Databases

TL;DR

This paper introduces Tiara, an efficient algorithm for users to retrieve optimal images from external databases using limited queries and black-box functions, enabling personalized image search.

Contribution

The paper proposes Tiara, a novel retrieval algorithm designed for users with limited access to external image databases, improving query efficiency for personalized image retrieval.

Findings

Tiara outperforms baseline methods in various experimental settings.

The method reduces the number of queries needed for optimal image retrieval.

Experiments confirm the effectiveness of Tiara in practical scenarios.

Abstract

Suppose we have a black-box function (e.g., deep neural network) that takes an image as input and outputs a value that indicates preference. How can we retrieve optimal images with respect to this function from an external database on the Internet? Standard retrieval problems in the literature (e.g., item recommendations) assume that an algorithm has full access to the set of items. In other words, such algorithms are designed for service providers. In this paper, we consider the retrieval problem under different assumptions. Specifically, we consider how users with limited access to an image database can retrieve images using their own black-box functions. This formulation enables a flexible and finer-grained image search defined by each user. We assume the user can access the database through a search query with tight API limits. Therefore, a user needs to efficiently retrieve optimal images in terms of the number of queries. We propose an efficient retrieval algorithm Tiara for this problem. In the experiments, we confirm that our proposed method performs better than several baselines under various settings.