Exploring Image Search on Quantum Computing Systems

Hermann Fürntratt, Werner Bailer, Florian Krebs, Roland Unterberger, Herwig Zeiner

2025

Abstract

Image search based on descriptor similarities is a fundamental task in computer vision. These descriptors are often highly quantised (e.g. binarised or ternarised). Quantum Computing (QC) has shown to have great potential for a number of tasks including search. Especially Google’s new Quantum Processing Unit called Willow reaches new milestones in error correction and durability and presents a benchmark that runs about five minutes in the quantum domain while it would take more than 1025 years on a state-of-the-art classical supercomputer. The supposed reason that this achievement is not immediately claimed for quantum supremacy is that Google also states that the benchmark — the so called Random Circuit Sampling (RCS) benchmark has ”not yet known real-world applications”. We therefore transform these benchmark results into the real world by exploring the question: what hardware specifications are needed to execute a quantum implementation of an image descriptor search algorithm more quickly than the fastest supercomputer available today? Hence, two distinct implementations of the Compact Descriptor for Video Applications (CDVA) search, which uses 512-bit descriptors are compared to classical code running on a MI300A unit available in El Capitan, the currently fastest supercomputer with AVX512-bit comands. The results indicate that the current key hardware factors, including gate runtime, coherence time, error rate, and the number of qubits, are still considerably (by orders of magnitude) below the performance levels necessary to compete with the El Capitan supercomputer, but with the recent progress in error correction, we can expect the development of larger quantum systems in the near future that reduces the performance gap between classical and quantum computers. Our source code for simulation is available online at GitHub.

Paper Citation

in Harvard Style

Fürntratt H., Bailer W., Krebs F., Unterberger R. and Zeiner H. (2025). Exploring Image Search on Quantum Computing Systems. In Proceedings of the 1st International Conference on Quantum Software - Volume 1: IQSOFT; ISBN 978-989-758-761-0, SciTePress, pages 70-78. DOI: 10.5220/0013562200004525

in Bibtex Style

@conference{iqsoft25,
author={Hermann Fürntratt and Werner Bailer and Florian Krebs and Roland Unterberger and Herwig Zeiner},
title={Exploring Image Search on Quantum Computing Systems},
booktitle={Proceedings of the 1st International Conference on Quantum Software - Volume 1: IQSOFT},
year={2025},
pages={70-78},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013562200004525},
isbn={978-989-758-761-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Quantum Software - Volume 1: IQSOFT
TI - Exploring Image Search on Quantum Computing Systems
SN - 978-989-758-761-0
AU - Fürntratt H.
AU - Bailer W.
AU - Krebs F.
AU - Unterberger R.
AU - Zeiner H.
PY - 2025
SP - 70
EP - 78
DO - 10.5220/0013562200004525
PB - SciTePress