Multi-view Drone-based Geo-localization via Style and Spatial Alignment

Abstract

In this paper, we focus on the task of multi-view multi-source geolocalization, which serves as an important auxiliary method of GPS positioning by matching drone-view image and satellite-view image with pre-annotated GPS tag. To solve this problem, most existing methods adopt metric loss with an weighted classification block to force the generation of common feature space shared by different view points and view sources. However, these methods fail to pay sufficient attention to spatial information (especially viewpoint variances). To address this drawback, we propose an elegant orientation-based method to align the patterns and introduce a new branch to extract aligned partial feature. Moreover, we provide a style alignment strategy to reduce the variance in image style and enhance the feature unification. To demonstrate the performance of the proposed approach, we conduct extensive experiments on the large-scale benchmark dataset. The experimental results confirm the superiority of the proposed approach compared to state-of-the-art alternatives.

A cross-view UAV image localization method based on directional feature alignment and visual transformer

“…In other words, using context information without using additional estimators. Hu et al [7] took into account the style deviation caused by camera, weather and seasonal changes, and adopted the method based on color levels to unify the style of UAV image and satellite image. In addition, they also used mesh division to carry out local feature alignment.…”