Search papers, labs, and topics across Lattice.
This paper introduces the Retrieval-In-Matching (RIM) framework, which enhances global visual localization for UAVs by addressing cross-domain appearance and viewpoint shifts through a two-stage fine-tuning process and local geometric consistency re-ranking. By utilizing a single DINOv2-B forward pass for both retrieval and local description, RIM achieves significant improvements in Recall@1 metrics over existing methods while maintaining high speed and efficiency. The results demonstrate RIM's capability to outperform ten recent retrieval baselines, establishing it as a robust solution for UAV localization in GNSS-challenged environments.
RIM achieves up to 13.77 percentage points improvement in localization recall while being over 40 times faster than traditional methods.
Global visual localization of unmanned aerial vehicles (UAVs) using remote-sensing reference maps has attracted increasing attention. However, acquisition-time and imaging-platform differences between UAV and reference imagery induce substantial cross-domain appearance and viewpoint shifts, challenging robust six-degree-of-freedom (6-DoF) pose estimation. We address these shifts by sampling UAV-viewpoint reference views from Google 3D Tiles across locations, altitudes, and orientations. A two-stage cross-domain fine-tuning recipe adapts SALAD using pose-near positives and geographically distant hard negatives, while local geometric consistency re-ranks the Top-K candidates. We further propose Retrieval-In-Matching (RIM), which freezes the adapted DINOv2-B retriever and distils a local-descriptor decoder that reuses its token field alongside a shallow VGG19 detail stream. One query-side DINOv2-B forward thus serves both SALAD retrieval and local description, eliminating a second foundation-model backbone while preserving retrieval descriptors by construction. We evaluate RIM zero-shot on the reconstructed EPFL Urbanscape and self-collected Chang'an Park datasets, both geographically disjoint from the training data. RIM outperforms ten recent retrieval baseline families. At 25/50 m under the full 3D distance metric, it improves Recall@1 over SALAD by 8.55/13.77 percentage points on EPFL and 4.45/8.94 points on Park. At Top-K=5, the complete measured localization query, including retrieval, candidate matching, and robust geometric verification, takes 67.9 ms end-to-end: 1.8 times faster than the strongest separate sparse-matching baseline and over 40 times faster than RoMa, while achieving comparable re-ranking accuracy. These results establish an efficient and deployable pipeline for UAV global visual localization in GNSS-challenged environments.
