Datasets › GSV-Cities

GSV-Cities

Introduced by Amar Ali-bey et al. in GSV-Cities: Toward Appropriate Supervised Visual Place Recognition19 Oct 2022 archive 2025-07-28

GSV-Cities is a large-scale dataset for training deep neural network for the task of Visual Place Recognition.

The dataset contains more than 530k images:

There are more than 62k different places, spread across multiple cities around the globe.
Each place is depited by at least 4 images (up to 20 images).
All places are physically distant (at least 100 meters between any pair of places).

Benchmarks archive 2025-07-28

No leaderboard in the archive resolves to this dataset.

No paper in the archive has a leaderboard row on this dataset; the archive counts 16 papers for it but never published that list.

1 loader as listed in the archive; links are outbound and not re-checked here.

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

No language tagged.

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections