Events

CS Special Seminar: Michal Piovarci "Breaking Constraints of Fabrication Hardware with Perception and Optimization"

This talk is arranged at the Department of Computer Science.
SpecialSeminar_AaltoEvent

Breaking Constraints of Fabrication Hardware with Perception and Optimization

Michal Piovarci
Institute for Science and Technology Austria
Homepage

Abstract: Faithfully reproducing real-world objects on a 3D printer is a challenging endeavor. Current 3D printers are constrained in terms of resolution and available materials to a relatively small set of fabricable artifacts. Moreover, oftentimes the goal of fabrication is unclear. While we possess an intuitive understanding of what we want to create providing a quantitative description suitable for fabrication is challenging. Interestingly, many of the attributes we seek to reproduce are closely tied to our senses. We utilize touch to perceive the fine geometrical structure, or our eyes to judge appearance. Similarly to fabrication hardware, our senses have inborn limitations given by biological constraints. For example, eyes have limited capability to distinguish high-frequency information; fingers feel applied forces in a non-linear fashion. In this talk, I will show you how to systematically investigate the human sensorial system to build its numerical approximations. We will then leverage the perceptual limitations to mask the constraints of fabrication hardware. In several applications, I will demonstrate that using perception and optimization it is possible to break before assumed fabrication constraints and produce high-quality replicas of real-world objects that were considered unprintable.

Bio: Michal Piovarci is a postdoctoral researcher at the Institute of Science and Technology in Austria. He is a member of the Digital Fabrication and Design group led by prof. Bernd Bickel. Michal obtained his Ph.D. at USI Lugano under the supervision of prof. Piotr Didyk. The key motivation of his thesis entitled Perception-Aware Computational Fabrication is to leverage human perception in the design of digital artifacts. Michal proposes a methodology for studying the human sensorial system and distilling its numerical approximations. He then demonstrates that incorporating perception has tangible benefits in creating high-quality replicas of haptic experiences indistinguishable from reality. The thesis received the prestigious Eurographics Ph.D. award and is recognized as seminal work in combining perception with fabrication by his peers. Currently, he is expanding the methodology towards appearance fabrication and optimal control of fabrication systems.

Department of Computer Science

To foster future science and society.

Read more
Mahine Learning researchers working at Department of Computer Science in Aalto University
  • Published:
  • Updated: