Both sides previous revision
Previous revision
Next revision
|
Previous revision
|
members:stumberg [2020/06/02 13:09] Lukas von Stumberg |
members:stumberg [2023/12/22 11:02] (current) Quirin Lohr |
<member> | <member> |
<name>Lukas von Stumberg</name> | <name>Lukas von Stumberg</name> |
<mail>stumberg@in.tum.de</mail> | <mail>LStumberg@outlook.com</mail> |
<title>PhD Student</title> | |
</member> | </member> |
Find me on [[https://scholar.google.de/citations?user=jBgFEukAAAAJ&hl=en&oi=ao|Google Scholar]] [[https://www.linkedin.com/in/lukas-von-stumberg-944255161|Linkedin]] | Find me on [[https://scholar.google.de/citations?user=jBgFEukAAAAJ&hl=en&oi=ao|Google Scholar]] [[https://www.linkedin.com/in/lukas-von-stumberg-944255161|Linkedin]] [[https://github.com/lukasvst|GitHub]] |
| |
==== Research Interests ==== | ==== Research Interests ==== |
| |
==== Selected Projects ==== | ==== Selected Projects ==== |
=== GN-Net: The Gauss-Newton Loss for Multi-Weather Relocalization === | === DM-VIO: Delayed Marginalization Visual-Inertial Odometry === |
Using our novel Gauss-Newton loss we improve the robustness of direct methods against strong illumination changes and bad initializations. This allows us to accurately relocalize between different weathers. [[:gn-net|Project Page]] | An open-source visual-inertial odometry system with state-of-the art results. |
| See https://github.com/lukasvst/dm-vio |
| |
<html> | <html> |
<iframe width="560" height="315" src="https://www.youtube.com/embed/q_uVb_o255o" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> | <iframe width="560" height="315" src="https://www.youtube.com/embed/7iep3BvcJPU" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> |
</html> | </html> |
| |
<html> | <html> |
| <iframe width="560" height="315" src="https://www.youtube.com/embed/pWKN7Afoirs" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> |
| </html> |
| |
| === LM-Reloc: Levenberg-Marquardt Based Direct Visual Relocalization === |
| As a follow-up of our previous work GN-Net, LM-Reloc boosts the performance of direct image alignment for relocalization using a novel loss formulation, and a pose estimation network. |
| <html> |
| <iframe width="560" height="315" src="https://www.youtube.com/embed/i7TyTwKD734" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> |
| </html> |
| |
| === GN-Net: The Gauss-Newton Loss for Multi-Weather Relocalization === |
| Using our novel Gauss-Newton loss we improve the robustness of direct methods against strong illumination changes and bad initializations. This allows us to accurately relocalize between different weathers. [[:gn-net|Project Page]] |
| |
| <html> |
| <iframe width="560" height="315" src="https://www.youtube.com/embed/q_uVb_o255o" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> |
| <br/> |
<iframe width="560" height="315" src="https://www.youtube.com/embed/gcbKeKX2eiE" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe></br></br> | <iframe width="560" height="315" src="https://www.youtube.com/embed/gcbKeKX2eiE" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe></br></br> |
</html> | </html> |