CloudTrack: Scalable UAV Tracking with Cloud Semantics

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Blei, Yannik, Krawez, Michael, Nilavadi, Nisarga, Kaiser, Tanja Katharina, Burgard, Wolfram
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911348770734080
author Blei, Yannik
Krawez, Michael
Nilavadi, Nisarga
Kaiser, Tanja Katharina
Burgard, Wolfram
author_facet Blei, Yannik
Krawez, Michael
Nilavadi, Nisarga
Kaiser, Tanja Katharina
Burgard, Wolfram
contents Nowadays, unmanned aerial vehicles (UAVs) are commonly used in search and rescue scenarios to gather information in the search area. The automatic identification of the person searched for in aerial footage could increase the autonomy of such systems, reduce the search time, and thus increase the missed person's chances of survival. In this paper, we present a novel approach to perform semantically conditioned open vocabulary object tracking that is specifically designed to cope with the limitations of UAV hardware. Our approach has several advantages. It can run with verbal descriptions of the missing person, e.g., the color of the shirt, it does not require dedicated training to execute the mission and can efficiently track a potentially moving person. Our experimental results demonstrate the versatility and efficacy of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2409_16111
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CloudTrack: Scalable UAV Tracking with Cloud Semantics
Blei, Yannik
Krawez, Michael
Nilavadi, Nisarga
Kaiser, Tanja Katharina
Burgard, Wolfram
Robotics
Computer Vision and Pattern Recognition
Nowadays, unmanned aerial vehicles (UAVs) are commonly used in search and rescue scenarios to gather information in the search area. The automatic identification of the person searched for in aerial footage could increase the autonomy of such systems, reduce the search time, and thus increase the missed person's chances of survival. In this paper, we present a novel approach to perform semantically conditioned open vocabulary object tracking that is specifically designed to cope with the limitations of UAV hardware. Our approach has several advantages. It can run with verbal descriptions of the missing person, e.g., the color of the shirt, it does not require dedicated training to execute the mission and can efficiently track a potentially moving person. Our experimental results demonstrate the versatility and efficacy of our approach.
title CloudTrack: Scalable UAV Tracking with Cloud Semantics
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.16111