Radu Alexandru Rosu

Autonomous Intelligent Systems Group

Address: 

   Radu Alexandru Rosu
   Rheinische Friedrich-Wilhelms-Universität Bonn
   Institut für Informatik VI
   Friedrich-Ebert-Allee 144
   53113 Bonn 

Tel:  +49 (0) 228 73-54159
Fax: +49 (0) 228 73-4425 

Email: rosu _at_ ais.uni-bonn.de

Office: 0.050

Curriculum vitae

---

Software Releases:

EasyPBR - Lightweight Physically-Based Renderer designed for ease of use and fast prototyping

EasyPBR is a lightweight Physically-Based Renderer designed for ease of use and fast prototyping. It abstracts away most of the complexity of rendering allowing the user to display high quality images in only a couple of lines of code. By default, EasyPBR tries to choose parameters in order to best render a particular scene but if more control is needed, the whole rendering process is controllable through either a Python script or C++ program. EasyPBR is written in C++ and OpenGL and provides Python bindings to most of it's functionality.

LatticeNet - Fast Point Cloud Segmentation Using Permutohedral Lattices

LatticeNet is a 3D neural network that can process raw point clouds for the task of semantic segmentation. Sparse permutohedral lattices are used to embedd the point cloud and apply convolutions in the high dimensional space of the lattice. We define novel operations that allow for data-dependent interpolation between the lattice and the point cloud. The implementation is written in C++ and CUDA and exposed through PyTorch.

---

Publications:

2020

• Radu Alexandru Rosu, Peer Schütt, Jan Quenzel, and Sven Behnke:
  LatticeNet: Fast Point Cloud Segmentation Using Permutohedral Lattices
  Accepted for Robotics: Science and Systems (RSS), July 2020.
  
  
• Jan Quenzel, Radu Alexandru Rosu, Thomas Läbe, Cyrill Stachniss, and Sven Behnke:
  Beyond Photometric Consistency: Gradient-based Dissimilarity for Improving Visual Odometry and Stereo Matching
  Accepted for IEEE International Conference on Robotics and Automation (ICRA), Paris, France, May 2020.
  
  

2019

• Julian Tanke, Oh-Hun Kwon, Patrick Stotko, Radu Alexandru Rosu, Michael Weinmann, Hassan Errami, Sven Behnke, Maren Bennewitz, Reinhard Klein, Andreas Weber, Angela Yao, Juergen Gall:
  Bonn Activity Maps: Dataset Description
  
  
• Radu Alexandru Rosu, Jan Quenzel, and Sven Behnke:
  Reconstruction of Textured Meshes for Fire and Heat Source Detection
  In Proceedings of IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), Würzburg, Germany, September 2019.
  
  
• Radu Alexandru Rosu, Jan Quenzel, and Sven Behnke:
  Semi-Supervised Semantic Mapping through Label Propagation with Semantic Texture Meshes
  In International Journal of Computer Vision (IJCV), Springer, 2019.
  
  
• Marius Beul, Matthias Nieuwenhuisen, Jan Quenzel, Radu Alexandru Rosu, Jannis Horn, Dmytro Pavlichenko, Sebastian Houben, and Sven Behnke:
  Team NimbRo at MBZIRC 2017: Fast Landing on a Moving Target and Treasure Hunting with a Team of MAVs
  Journal of Field Robotics (JFR) 36(1):204-229, Wiley, January 2019.
  [Video]
  
  

2017

• Jan Quenzel, Radu Alexandru Rosu, Sebastian Houben, and Sven Behnke:
  Online Depth Calibration for RGB-D Cameras using Visual SLAM
  In Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, Canada, September 2017.
  
  
• Matthias Nieuwenhuisen, Marius Beul, Radu Alexandru Rosu, Jan Quenzel, Dmytro Pavlichenko, Sebastian Houben, and Sven Behnke:
  Collaborative Object Picking and Delivery with a Team of Micro Aerial Vehicles at MBZIRC
  In Proceedings of European Conference on Mobile Robotics (ECMR), Paris, France, September 2017.