EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Application of a novel deep learning–based 3D videography workflow to bat flight

Jonas Håkansson iD, Brooke L Quinn, Abigail L Shultz, Sharon M Swartz, Aaron J Corcoran

Annals of the New York Academy of Sciences · 2024

DOI: 10.1111/nyas.15143 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Annals of the New York Academy of Sciences, 2024. Verfasst von Jonas Håkansson und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biology, University of Colorado Colorado Springs, Colorado Springs, Colorado, USA. Hauptthemen laut MEDLINE: Chiroptera, Flight, Animal, Deep Learning, Workflow. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Studying the detailed biomechanics of flying animals requires accurate three‐dimensional coordinates for key anatomical landmarks. Traditionally, this relies on manually digitizing animal videos, a labor‐intensive task that scales poorly with increasing framerates and numbers of cameras. Here, we present a workflow that combines deep learning–powered automatic digitization with filtering and correction of mislabeled points using quality metrics from deep learning and 3D reconstruction. We tested our workflow using a particularly challenging scenario: bat flight. First, we documented four bats flying steadily in a 2 m 3 wind tunnel test section. Wing kinematic parameters resulting from manually digitizing bats with markers applied to anatomical landmarks were not significantly different from those resulting from applying our workflow to the same bats without markers for five out of six parameters. Second, we compared coordinates from manual digitization against those yielded via our workflow for bats flying freely in a 344 m 3 enclosure. Average distance between coordinates from our workflow and those from manual digitization was less than a millimeter larger than the average human‐to‐human coordinate distance. The improved efficiency of our workflow has the potential to increase the scalability of studies on animal flight biomechanics.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ChiropteraFlight, AnimalDeep LearningWorkflowVideo RecordingImaging, Three-DimensionalAnimalsBiomechanical PhenomenaWings, Animal

Stichwörter der Autor:innen

3D reconstructionanimal flightbiomechanicsdeep learningkinematicsvideography

Autor:innen und Institutionen

  • Department of Biology, University of Colorado Colorado Springs, Colorado Springs, Colorado, USA.
  • Department of Ecology, Evolution, and Organismal Biology, Brown University, Providence, Rhode Island, USA.
  • School of Engineering, Brown University, Providence, Rhode Island, USA.

ORCID-Profile: Jonas Håkansson

Förderung

  • National Science Foundation

Zitieren

APA

Håkansson, J., Quinn, B. L., Shultz, A. L., Swartz, S. M., & Corcoran, A. J. (2024). Application of a novel deep learning–based 3D videography workflow to bat flight. Annals of the New York Academy of Sciences. https://doi.org/10.1111/nyas.15143

Vancouver

Håkansson J, Quinn BL, Shultz AL, Swartz SM, Corcoran AJ. Application of a novel deep learning–based 3D videography workflow to bat flight. Annals of the New York Academy of Sciences. 2024. doi:10.1111/nyas.15143

BibTeX
@article{Hkansson2024,
  author = {Håkansson, Jonas and Quinn, Brooke L and Shultz, Abigail L and Swartz, Sharon M and Corcoran, Aaron J},
  title = {Application of a novel deep learning–based 3D videography workflow to bat flight},
  journal = {Annals of the New York Academy of Sciences},
  year = {2024},
  doi = {10.1111/nyas.15143}
}
Kostenlos starten →