LoD-Loc v3: Generalized Aerial Localization in Dense Cities using Instance Silhouette Alignment
This work addresses localization challenges for applications like drones or autonomous systems in dense cities, representing a significant but incremental advance over prior methods.
The paper tackles the problem of aerial visual localization in dense urban environments by introducing LoD-Loc v3, which uses instance silhouette alignment and a large synthetic dataset to improve cross-scene generalization and performance in dense building scenes, achieving superior results over state-of-the-art baselines with a large margin.
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/.