Random Binary Trees for Approximate Nearest Neighbour Search in Binary Space

TL;DR

This paper introduces a novel search method using Random Binary Search Trees for high-dimensional binary vectors, demonstrating superior retrieval precision over existing LSH methods in real-world image localization data.

Contribution

The paper proposes a simple, effective search algorithm based on RBSTs for high-dimensional binary data, outperforming state-of-the-art LSH variants in accuracy.

Findings

RBST outperforms LSH variants in retrieval precision

Performance boost of over 20% in real-world datasets

Effective for high-dimensional binary feature search

Abstract

Approximate nearest neighbour (ANN) search is one of the most important problems in computer science fields such as data mining or computer vision. In this paper, we focus on ANN for high-dimensional binary vectors and we propose a simple yet powerful search method that uses Random Binary Search Trees (RBST). We apply our method to a dataset of 1.25M binary local feature descriptors obtained from a real-life image-based localisation system provided by Google as a part of Project Tango. An extensive evaluation of our method against the state-of-the-art variations of Locality Sensitive Hashing (LSH), namely Uniform LSH and Multi-probe LSH, shows the superiority of our method in terms of retrieval precision with performance boost of over 20%