2023 edition: Tasks | Methodology | Datasets | Repository examples | Committee and pre-registration

SISAP 2023 Indexing Challenge and Demo Track

The International Conference on Similarity Search and Applications (SISAP) is an annual forum for researchers and application developers in the area of similarity data management. It aims at the technological problems shared by numerous application domains, such as data mining, information retrieval, multimedia, computer vision, pattern recognition, computational biology, geography, biometrics, machine learning, and many others that use similarity search as a necessary supporting service. Machine learning, dense retrieval, and multimedia indexing have taken the scene of similarity search as the most challenging task among applications. This call reflects on this observation.

In 2023 we are launching the SISAP Indexing Challenge with a test bed to compare new and existing indexing algorithms in three common tasks. Current data pools are in the order of several billion objects to train a generative deep neural network; we will be steering towards indexing datasets in this size in upcoming editions of this challenge. For this year, we are launching the challenge with datasets fitting on a single modern desktop computer, sized from ten million to one hundred million objects. We aim at providing a common test bed to compare new indexes, re-implementations of previously known indexes and an overview of the behavior of similarity search at various size scales.

Important information


[1] Webster, R., Rabin, J., Simon, L., & Jurie, F. (2023). On the De-duplication of LAION-2B. arXiv preprint arXiv:2303.12733.

Important dates

[2] Visit committee for more instructions.


We expect that participants prepare a detailed report of their solution in a typical SISAP's shortpaper format with a focus on reproducibility and comparing their speedup against a brute force solution in its machine and the resulting quality w.r.t. recall. Organizers and an ad-hoc committee will review these short papers if the number of papers is large, looking to increase the paper's quality. All participants can submit a report regardless of its final rank position. As in other SISAP's tracks and sessions, the acceptance will be based on the quality of contribution and the manuscript itself.

CC BY-SA 4.0 sisap challenge committee. Last modified: April 08, 2024. Website built with Franklin.jl and the Julia programming language.