Salvato in:
Dettagli Bibliografici
Autori principali: Tan, Zhuoyue, He, Boyong, Ji, Yuxiang, Wu, Liaoni
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2506.14525
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912436247855104
author Tan, Zhuoyue
He, Boyong
Ji, Yuxiang
Wu, Liaoni
author_facet Tan, Zhuoyue
He, Boyong
Ji, Yuxiang
Wu, Liaoni
contents This paper presents VisLanding, a monocular 3D perception-based framework for safe UAV (Unmanned Aerial Vehicle) landing. Addressing the core challenge of autonomous UAV landing in complex and unknown environments, this study innovatively leverages the depth-normal synergy prediction capabilities of the Metric3D V2 model to construct an end-to-end safe landing zones (SLZ) estimation framework. By introducing a safe zone segmentation branch, we transform the landing zone estimation task into a binary semantic segmentation problem. The model is fine-tuned and annotated using the WildUAV dataset from a UAV perspective, while a cross-domain evaluation dataset is constructed to validate the model's robustness. Experimental results demonstrate that VisLanding significantly enhances the accuracy of safe zone identification through a depth-normal joint optimization mechanism, while retaining the zero-shot generalization advantages of Metric3D V2. The proposed method exhibits superior generalization and robustness in cross-domain testing compared to other approaches. Furthermore, it enables the estimation of landing zone area by integrating predicted depth and normal information, providing critical decision-making support for practical applications.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14525
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VisLanding: Monocular 3D Perception for UAV Safe Landing via Depth-Normal Synergy
Tan, Zhuoyue
He, Boyong
Ji, Yuxiang
Wu, Liaoni
Computer Vision and Pattern Recognition
Robotics
This paper presents VisLanding, a monocular 3D perception-based framework for safe UAV (Unmanned Aerial Vehicle) landing. Addressing the core challenge of autonomous UAV landing in complex and unknown environments, this study innovatively leverages the depth-normal synergy prediction capabilities of the Metric3D V2 model to construct an end-to-end safe landing zones (SLZ) estimation framework. By introducing a safe zone segmentation branch, we transform the landing zone estimation task into a binary semantic segmentation problem. The model is fine-tuned and annotated using the WildUAV dataset from a UAV perspective, while a cross-domain evaluation dataset is constructed to validate the model's robustness. Experimental results demonstrate that VisLanding significantly enhances the accuracy of safe zone identification through a depth-normal joint optimization mechanism, while retaining the zero-shot generalization advantages of Metric3D V2. The proposed method exhibits superior generalization and robustness in cross-domain testing compared to other approaches. Furthermore, it enables the estimation of landing zone area by integrating predicted depth and normal information, providing critical decision-making support for practical applications.
title VisLanding: Monocular 3D Perception for UAV Safe Landing via Depth-Normal Synergy
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2506.14525