EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SGFAM: Semantic and Geometric Features Aggregation for Dense Shape Matching in Generalizable Robotic Manipulation

Paolo Sebeto, Christian Hartl-Nesic, Jean-Baptiste Weibel, Daniel Zimmer, Andreas Holzinger, Markus Vincze

IEEE Robotics and Automation Letters · 2026-06 · 11 (6) · 7254-7261

DOI: 10.1109/lra.2026.3685940 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Robotics and Automation Letters, 2026-06. Verfasst von Paolo Sebeto und 5 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: EC Horizon 2020 for Research and Innovation Project TraceBot, Open Access Funding Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To automate processes like polishing or cleaning at scale, robots must be able to adapt learned skills to new object instances without manual reprogramming. Applications requiring tool-surface interactions face a significant challenge in transferring manipulation strategies to novel objects due to substantial shape and appearance variations. Robust, generalized dense shape correspondence is essential for solving this problem. We present SGFAM, a zero-shot pipeline integrating pre-trained vision foundation models and geometric encoders via functional maps. Unlike prior works that rely on simple averaging, we introduce 1) an alignment-based feature aggregation to prioritize optimal viewing angles, and 2) Kernel PCA fusion to preserve non-linear descriptor manifolds. Our evaluations demonstrate that this approach not only outperforms state-of-the-art baselines but also enables lightweight vision backbones to achieve matching precision comparable to larger models. We validate SGFAM experimentally by successfully transferring continuous surface paths in real-world industrial and household robotic scenarios without requiring any finetuning.

Förderung

  • EC Horizon 2020 for Research and Innovation Project TraceBot
  • Open Access Funding Program

Zitieren

APA

Sebeto, P., Hartl-Nesic, C., Weibel, J., Zimmer, D., Holzinger, A., & Vincze, M. (2026). SGFAM: Semantic and Geometric Features Aggregation for Dense Shape Matching in Generalizable Robotic Manipulation. IEEE Robotics and Automation Letters, 11(6), 7254-7261. https://doi.org/10.1109/lra.2026.3685940

Vancouver

Sebeto P, Hartl-Nesic C, Weibel J, Zimmer D, Holzinger A, Vincze M. SGFAM: Semantic and Geometric Features Aggregation for Dense Shape Matching in Generalizable Robotic Manipulation. IEEE Robotics and Automation Letters. 2026;11(6):7254-7261. doi:10.1109/lra.2026.3685940

BibTeX
@article{Sebeto2026,
  author = {Sebeto, Paolo and Hartl-Nesic, Christian and Weibel, Jean-Baptiste and Zimmer, Daniel and Holzinger, Andreas and Vincze, Markus},
  title = {SGFAM: Semantic and Geometric Features Aggregation for Dense Shape Matching in Generalizable Robotic Manipulation},
  journal = {IEEE Robotics and Automation Letters},
  year = {2026},
  volume = {11},
  number = {6},
  pages = {7254--7261},
  doi = {10.1109/lra.2026.3685940}
}
Kostenlos starten →