GeoDTR+: Toward generic cross-view geolocalization via geometric disentanglement

TL;DR

GeoDTR+ introduces an enhanced geometric layout extractor and contrastive hard sample generation to improve cross-area geolocalization, achieving state-of-the-art results by better modeling visual feature correlations and layout.

Contribution

The paper proposes GeoDTR+ with an improved GLE module and CHSG technique, significantly enhancing cross-area geolocalization performance over previous methods.

Findings

Achieves SOTA results on CVUSA, CVACT, and VIGOR datasets.

Improves cross-area evaluation accuracy by over 13-22%.

Maintains comparable performance in same-area settings.

Abstract

Cross-View Geo-Localization (CVGL) estimates the location of a ground image by matching it to a geo-tagged aerial image in a database. Recent works achieve outstanding progress on CVGL benchmarks. However, existing methods still suffer from poor performance in cross-area evaluation, in which the training and testing data are captured from completely distinct areas. We attribute this deficiency to the lack of ability to extract the geometric layout of visual features and models' overfitting to low-level details. Our preliminary work introduced a Geometric Layout Extractor (GLE) to capture the geometric layout from input features. However, the previous GLE does not fully exploit information in the input feature. In this work, we propose GeoDTR+ with an enhanced GLE module that better models the correlations among visual features. To fully explore the LS techniques from our preliminary work, we further propose Contrastive Hard Samples Generation (CHSG) to facilitate model training. Extensive experiments show that GeoDTR+ achieves state-of-the-art (SOTA) results in cross-area evaluation on CVUSA, CVACT, and VIGOR by a large margin ($16.44\%$, $22.71\%$, and $13.66\%$ without polar transformation) while keeping the same-area performance comparable to existing SOTA. Moreover, we provide detailed analyses of GeoDTR+. Our code will be available at https://gitlab.com/vail-uvm/geodtr plus.