MF-UAVPose6D: A Model-Free Monocular 6-DoF Pose Estimation Framework for Fixed-Wing UAVs
2026-06-29 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionRobotics
AI summaryⓘ
The authors developed a way to figure out the exact position and orientation (6-DoF pose) of unmanned airplanes using just one regular color photo, without needing detailed 3D models. They made a method that finds the plane in the image, understands its shape from different parts like wings and tail, and then calculates its 3D pose. They also created a synthetic dataset for testing. Their approach works well even when the plane is far away or seen from tricky angles.
6-DoF pose estimationuncrewed aerial vehicles (UAVs)monocular visionPerspective-Aware ModuleDynamic Topological Samplingcamera intrinsicssynthetic datasetfixed-wing UAVtranslation-rotation decodingheatmap localization
Authors
Juanqin Liu, Leonardo Plotegher, Eloy Roura, Shaoming He
Abstract
For uncrewed aerial vehicles (UAVs), estimating six-degree-of-freedom (6-DoF) poses is essential for airspace situational awareness, target tracking, and counter-UAV operations. However, non-cooperative targets usually lack computer-aided design (CAD) models and keypoint priors, making existing model-based or keypoint-matching methods difficult to apply reliably. To address these challenges, this paper proposes MF-UAVPose6D, a model-free monocular 6-DoF pose estimation framework for fixed-wing UAVs. During inference, the method takes only a single red-green-blue (RGB) image and camera intrinsics as input. It first obtains a stable target anchor through heatmap-guided center localization, introduces a Perspective-Aware Module (PAM) to model observation-ray priors, exploits Dynamic Topological Sampling (DTS) to complement weak structural cues from the wings, fuselage, and tail, and adopts a decoupled translation-rotation pose decoding mechanism to estimate the 6-DoF pose. In addition, we construct the FW-UAV6DPose synthetic dataset, which covers fixed-wing UAV observations across diverse distances, viewpoints, and poses. Experimental results show that MF-UAVPose6D achieves accurate and efficient monocular 6-DoF pose estimation without requiring CAD models, and demonstrates strong robustness in long-range rotation estimation, depth recovery, and joint pose evaluation.