Orientation-Guided Contrastive Learning for UAV-View Geo-Localisation

TL;DR

This paper introduces an orientation-guided contrastive learning framework for UAV-view geo-localisation, improving accuracy by estimating and utilizing orientation information without extra inference costs.

Contribution

It proposes a lightweight orientation prediction module and pseudo-labels to enhance contrastive learning for UAV geo-localisation, achieving state-of-the-art results.

Findings

Outperforms previous methods on University-1652 and University-160k datasets

Uses pseudo-labels for orientation to improve model training

No additional computation needed during inference

Abstract

Retrieving relevant multimedia content is one of the main problems in a world that is increasingly data-driven. With the proliferation of drones, high quality aerial footage is now available to a wide audience for the first time. Integrating this footage into applications can enable GPS-less geo-localisation or location correction. In this paper, we present an orientation-guided training framework for UAV-view geo-localisation. Through hierarchical localisation orientations of the UAV images are estimated in relation to the satellite imagery. We propose a lightweight prediction module for these pseudo labels which predicts the orientation between the different views based on the contrastive learned embeddings. We experimentally demonstrate that this prediction supports the training and outperforms previous approaches. The extracted pseudo-labels also enable aligned rotation of the satellite image as augmentation to further strengthen the generalisation. During inference, we no longer need this orientation module, which means that no additional computations are required. We achieve state-of-the-art results on both the University-1652 and University-160k datasets.