LoD-Loc v3: Generalized Aerial Localization in Dense Cities using Instance Silhouette Alignment

TL;DR

LoD-Loc v3 introduces a new aerial localization method that leverages a large synthetic dataset and instance silhouette alignment to improve accuracy and generalization in dense urban environments.

Contribution

It presents a synthetic dataset and a novel silhouette alignment approach that significantly enhance aerial localization in complex city scenes.

Findings

Outperforms existing SOTA methods in dense urban scenarios

Exhibits remarkable zero-shot generalization to new scenes

Achieves superior accuracy in pose estimation

Abstract

We present LoD-Loc v3, a novel method for generalized aerial visual localization in dense urban environments. While prior work LoD-Loc v2 achieves localization through semantic building silhouette alignment with low-detail city models, it suffers from two key limitations: poor cross-scene generalization and frequent failure in dense building scenes. Our method addresses these challenges through two key innovations. First, we develop a new synthetic data generation pipeline that produces InsLoD-Loc - the largest instance segmentation dataset for aerial imagery to date, comprising 100k images with precise instance building annotations. This enables trained models to exhibit remarkable zero-shot generalization capability. Second, we reformulate the localization paradigm by shifting from semantic to instance silhouette alignment, which significantly reduces pose estimation ambiguity in dense scenes. Extensive experiments demonstrate that LoD-Loc v3 outperforms existing state-of-the-art (SOTA) baselines, achieving superior performance in both cross-scene and dense urban scenarios with a large margin. The project is available at https://nudt-sawlab.github.io/LoD-Locv3/.