# EarthLoc: Astronaut Photography Localization by Indexing Earth from Space

Source: [EarthLoc: Astronaut Photography Localization by Indexing Earth from Space](<https://www.sota2.com/research/paper/68fa72751e209131cea567a4>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-02T04:59:17.457Z

Authors: Gabriele Berton, Alex Stoken, Barbara Caputo, Carlo Masone

Published: 2024-03-11T14:30:51.000Z

## Abstract

Astronaut photography, spanning six decades of human spaceflight, presents a unique Earth observations dataset with immense value for both scientific research and disaster response. Despite its significance, accurately localizing the geographical extent of these images, crucial for effective utilization, poses substantial challenges. Current manual localization efforts are time-consuming, motivating the need for automated solutions. We propose a novel approach - leveraging image retrieval - to address this challenge efficiently. We introduce innovative training techniques, including Year-Wise Data Augmentation and a Neutral-Aware Multi-Similarity Loss, which contribute to the development of a high-performance model, EarthLoc. We develop six evaluation datasets and perform a comprehensive benchmark comparing EarthLoc to existing methods, showcasing its superior efficiency and accuracy. Our approach marks a significant advancement in automating the localization of astronaut photography, which will help bridge a critical gap in Earth observations data. Code and datasets are available at https://github.com/gmberton/EarthLoc

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Visual Place Recognition on Texas (val)](<https://www.sota2.com/research/sota/visual-place-recognition-on-texas-val.md>): Visual Place Recognition; Texas (val); R@1: 54.6.

- [Visual Place Recognition on Alps (test)](<https://www.sota2.com/research/sota/visual-place-recognition-on-alps-test.md>): Visual Place Recognition; Alps (test); R@1: 53.9.

- [Visual Place Recognition on California (test)](<https://www.sota2.com/research/sota/visual-place-recognition-on-california-test.md>): Visual Place Recognition; California (test); Recall@1: 55.9.

- [Visual Place Recognition on Gobi Desert (test)](<https://www.sota2.com/research/sota/visual-place-recognition-on-gobi-desert-test.md>): Visual Place Recognition; Gobi Desert (test); R@1: 46.8.

- [Visual Place Recognition on Amazon (test)](<https://www.sota2.com/research/sota/visual-place-recognition-on-amazon-test.md>): Visual Place Recognition; Amazon (test); R@1: 45.6.

- [Visual Place Recognition on Toshka Lakes (test)](<https://www.sota2.com/research/sota/visual-place-recognition-on-toshka-lakes-test.md>): Visual Place Recognition; Toshka Lakes (test); R@1: 67.6.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2403.06758>)

[PDF](<https://arxiv.org/pdf/2403.06758.pdf>)

[Implementation](<https://github.com/gmberton/EarthLoc>)
