EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Sim-to-Real Dataset of Industrial Metal Objects

Peter De Roovere, Steven Moonen, Nick Michiels, Francis wyffels

Machines · 01.02.2024 · 12 (2) · 99

DOI: 10.3390/machines12020099 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machines, online erschienen am 01.02.2024. Verfasst von Peter De Roovere und 3 weiteren Autor:innen. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: VLAIO Baekeland Mandate, Flanders Make’s PILS SBO project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a diverse dataset of industrial metal objects with unique characteristics such as symmetry, texturelessness, and high reflectiveness. These features introduce challenging conditions that are not captured in existing datasets. Our dataset comprises both real-world and synthetic multi-view RGB images with 6D object pose labels. Real-world data were obtained by recording multi-view images of scenes with varying object shapes, materials, carriers, compositions, and lighting conditions. This resulted in over 30,000 real-world images. We introduce a new public tool that enables the quick annotation of 6D object pose labels in multi-view images. This tool was used to provide 6D object pose labels for all real-world images. Synthetic data were generated by carefully simulating real-world conditions and varying them in a controlled and realistic way. This resulted in over 500,000 synthetic images. The close correspondence between synthetic and real-world data and controlled variations will facilitate sim-to-real research. Our focus on industrial conditions and objects will facilitate research on computer vision tasks, such as 6D object pose estimation, which are relevant for many industrial applications, such as machine tending. The dataset and accompanying resources are available on the project website.

Förderung

  • VLAIO Baekeland Mandate
  • Flanders Make’s PILS SBO project

Zitieren

APA

Roovere, P. D., Moonen, S., Michiels, N., & wyffels, F. (2024). Sim-to-Real Dataset of Industrial Metal Objects. Machines, 12(2), 99. https://doi.org/10.3390/machines12020099

Vancouver

Roovere PD, Moonen S, Michiels N, wyffels F. Sim-to-Real Dataset of Industrial Metal Objects. Machines. 2024;12(2):99. doi:10.3390/machines12020099

BibTeX
@article{Roovere2024,
  author = {Roovere, Peter De and Moonen, Steven and Michiels, Nick and wyffels, Francis},
  title = {Sim-to-Real Dataset of Industrial Metal Objects},
  journal = {Machines},
  year = {2024},
  volume = {12},
  number = {2},
  pages = {99},
  doi = {10.3390/machines12020099}
}
Kostenlos starten →