# Learning Geo-Temporal Image Features

Source: [Learning Geo-Temporal Image Features](<https://www.sota2.com/research/paper/68fa66621e209131cea10011>)

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

Data updated: 2025-10-25T21:49:28.917Z

Authors: Menghua Zhai, Tawfiq Salem, Connor Greenwell, Scott Workman, Robert Pless, Nathan Jacobs

Published: 2019-09-16T22:00:59.000Z

## Abstract

We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks. To train our method, we take advantage of a large image dataset, captured by outdoor webcams and cell phones. The only form of supervision we provide are the known capture time and location of each image. We find that our approach learns features that are related to natural appearance changes in outdoor scenes. Additionally, we demonstrate the application of these geo-temporal features to time and location estimation.

## 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.

- [Time prediction on CVT (test)](<https://www.sota2.com/research/sota/time-prediction-on-cvt-test.md>): Time Prediction; CVT (test); ToY Error: 60.42.

- [Geo-time Aware Image Retrieval on TIGeR 86k (test)](<https://www.sota2.com/research/sota/geo-time-aware-image-retrieval-on-tiger-86k-test.md>): Geo-time Aware Image Retrieval; TIGeR 86k (test); Recall@1: 2.6.

- [Compositional Image Retrieval on TIGeR 86k (test)](<https://www.sota2.com/research/sota/compositional-image-retrieval-on-tiger-86k-test.md>): Compositional Image Retrieval; TIGeR 86k (test); Recall@1: 13.9.

- [Geo-localization on TIGeR 86k (test)](<https://www.sota2.com/research/sota/geo-localization-on-tiger-86k-test.md>): Geo-localization; TIGeR 86k (test); Recall@200km: 28.83.

- [Compositional Image Retrieval on CVT](<https://www.sota2.com/research/sota/compositional-image-retrieval-on-cvt.md>): Compositional Image Retrieval; CVT; Recall@1: 13.64.

- [Time prediction on TIGeR 86k (test)](<https://www.sota2.com/research/sota/time-prediction-on-tiger-86k-test.md>): Time Prediction; TIGeR 86k (test); ToY Error: 55.15.

- [Geo-time Aware Image Retrieval on CVT](<https://www.sota2.com/research/sota/geo-time-aware-image-retrieval-on-cvt.md>): Geo-time Aware Image Retrieval; CVT; R@1: 2.95.

- [Geo-localization on CVT (test)](<https://www.sota2.com/research/sota/geo-localization-on-cvt-test.md>): Geo-localization; CVT (test); Recall@200km: 23.82.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1909.07499>)

[PDF](<https://arxiv.org/pdf/1909.07499.pdf>)