Weibel, J. B., Patten, T., & Vincze, M. (2022). Robust Sim2Real 3D Object Classification Using Graph Representations and a Deep Center Voting Scheme. IEEE Robotics and Automation Letters, 7(3), 8028–8035. https://doi.org/10.1109/LRA.2022.3186745
deep learning for visual perception; recognition; visual learning
en
Abstract:
While object semantic understanding is essential for service robotic tasks, 3D object classification is still an open problem. Learning from artificial 3D models alleviates the cost of the annotation necessary to approach this problem, but today's methods still struggle with the differences between artificial and real 3D data. We conjecture that one of the causes of this issue is the fact that today's methods learn directly from point coordinates, which makes them highly sensitive to scale changes. We propose to learn from a graph of reproducible object parts whose scale is more reliable. In combination with a voting scheme, our approach achieves significantly more robust classification and improves upon state-of-the-art by up to 16% when transferring from artificial to real objects.
en
Project title:
Verfolgbare Roboter Handhabung von sterilen medizinischen Produkten: 101017089 (European Commission) Mensch geführtes Lernen um Haufen von Gegenständen zu sortieren: I 3968-N30 (Fonds zur Förderung der wissenschaftlichen Forschung (FWF))
