Nov 2026· Proceedings of the 4th International Workshop on UAVs in Multimedia: Capturing the World from a New Perspective· pp. 35-39· 1 citation· 6 references
TL;DR
LaMP (Last-meter Manifold Pose estimation) is proposed, a framework that builds on the manifold prior induced by the target-centric terminal approach and injects it through supervision rather than as a loss constraint, lifting the task from predicting its two scalar outputs, heading and range change, to estimating the full 6-DoF relative pose.
Abstract
Image-goal terminal UAV navigation guides a UAV from its current viewpoint to a goal viewpoint specified by a reference image. Local feature matching is brittle under the large viewpoint change between the two views, and monocular geometry alone struggles to supply the target-centric metric range the task demands. We propose LaMP (Last-meter Manifold Pose estimation), a framework that builds on the manifold prior induced by the target-centric terminal approach and injects it through supervision rather than as a loss constraint. Specifically, it lifts the task from predicting its two scalar outputs, heading and range change, to estimating the full 6-DoF relative pose, with labels constructed in closed form from the trajectory geometry. A geometry-supervised pose head reads heading from its continuous rotation, and a lightweight range head trained with a loss matched to the task’s relative-error metric supplements it for metric range. Both heads share a frozen VGGT backbone that encodes the two views jointly. LaMP attains sub-degree heading and sub-meter range error on held-out buildings and an official error of 0.0024 on the PairUAV hidden test (5th on the final leaderboard). Across unsampled viewpoints, altered approach slopes, and real drone imagery, heading remains accurate while metric range follows the assumed approach geometry. Code is publicly available at https://github.com/Polaris-F/vggt-pairuav.
In the image-driven UAV navigation task using the PairUAV dataset, we aim to estimate relative heading and signed translation range between a source-target image pair for last-meter pose alignment. This task faces challenges including drastic viewpoint/scale changes, repeated scenes, low-texture areas and periodic head...
Guo-Zheng Zheng, Ming-Jie Xie, Cong-Yi Fan et al.· Proceedings of the 4th Inter...· 1 citation
The results demonstrate that an ordinary globally localized ground vehicle can provide useful metric drift correction for monocular UAV vSLAM without artificial markers, UWB ranging, or additional onboard UAV localization sensors.
Károly Fodor, Márton Cserni, A. Rovid· IEEE Access· 0 citations
Last-meter UAV navigation requires predicting the relative pose between a source view and a target view under severe scale ambiguity and limited co-visibility, a regime in which cross-view retrieval and appearance-based matching pipelines struggle. By treating this fine-grained pose alignment as an inherent two-view ge...
Rui-Qi Yu, Zi-Qi Wang, Hai-Feng Xia et al.· Proceedings of the 4th Inter...· 1 citation
We present a per-pair UAV pose-estimation pipeline for orbital image sets built around a multi-modal fusion architecture: three frozen vision foundation models — DINOv2 visual semantics, SuperPoint+LightGlue correspondences, and DepthAnythingV2 depth statistics — are fused through a lightweight cross-attention head wit...
Xin-Yan Liu, Wei-Gang Zhang, Wei-Dong Chen et al.· Proceedings of the 4th Inter...· 1 citation
Vision and language navigation (VLN) for unmanned aerial vehicles (UAVs) requires an agent to follow natural language instructions while perceiving scenes whose appearance changes with flight altitude and heading. Existing methods treat these variations as nuisances to be learned, spending model capacity on scale and o...
Ergashevich Halimjon Khujamatov, Khasanov Doston, Yusupov Sarvarbek Sodikovich et al.· Italian National Conference...· 0 citations