Uni-Logo
You are here: Home Staff Noha Radwan
Articleactions

Noha Radwan

Albert-Ludwigs-Universität Freiburg
Technische Fakultät
Autonome Intelligente Systeme
Georges-Köhler-Allee 080
D-79110 Freiburg i. Br., Germany
Office:   080-00-008

radwann@informatik.uni-freiburg.de
Phone:   +49 761 203-8011
Fax:   +49 761 203-8007

About me

I am a postdoctoral researcher in the Autonomous Intelligent Systems group of the Department of Computer Science at the University of Freiburg in Germany.

My research is focused on addressing the challenges associated with localization and state estimation in urban environments using techniques from computer vision and machine learning, with the overall goal of enabling scalable, lifelong behavior.

Education

  • 09/2007 - 06/2012: Bachelor o.Sc. in Computer Science, German University in Cairo
  • 10/2012 - 03/2015: Master o.Sc. in Computer Science, University of Freiburg
  • 04/2015 - 06/2019: Dr. rer. nat. (Ph.D.) in Computer Science, Unversity of Freiburg
  • 06/2019 - up to now: Working at the Autonomous Intelligent Systems group of Prof. Dr. Wolfram Burgard as Research Assistant.

Research Interests

  • Robot Perception
  • Visual Localization
  • Robot Learning

Teaching

Students Supervised

  • Dynamic Object Invariant Space Recovery, Borna Bešić, 2018, Master's Project (Ongoing)
  • Landmark-based Visual Localization using Deep Convolutional Neural Networks, Jay Patravali, 2017, Internship
  • Multimodal Localization using Deep Convolutional Neural Networks, Hanna Stellmach, 2017, Master's Project
  • Text Spotting using Attention Models, Claas Bollen, 2017, Master's Project
  • Text Recognition in Urban Environments, Larissa Ho, 2016, Internship

Videos

News
  • Jun 2019: I will serve as the Program Committee Co-Chair of RSS Pioneers 2020.
  • Jun 2019: I have successfully defended my PhD thesis.
  • April 2019: I have been selected for RSS Pioneers which will be held in conjunction with Robotics: Science and Systems 2019
  • Feb-May 2019: Interning at Google Brain/Waymo
  • Sept 2018: VLocNet++ which achieves state-of-the-art visual localization performance by leveraging the semantic and geometric knowledge of the environment is accepted for IEEE RA-L journal.
  • Jul 2018: Our workshop paper on Effective Interaction-aware Trajectory Prediction using Temporal Convolutional Neural Networks got accepted for IROS'18 workshop on Crowd Navigation.
  • Jun 2018: Our workshop paper on Incorporating Semantic and Geometric Priors in Deep Pose Regression got accepted for RSS'18-LAIR.
  • May 2018: DeepLoc goes live!
  • May 2018: Our proposal on Autonomous Street Crossing with City Navigation Robots just got accepted by the DFG.
  • Jan 2018: VLocNet the first CNN based localization method to outperform local feature based techniques is accepted for ICRA'18.
  • Nov 2017: I was awarded the Doctoral Consortium Grant at ISRR'17.
  • Nov 2017: Invited to give a talk titled Towards Integrating Autonomous Robots in Urban Environments at the Observing Leaders of the future session at World Youth Forum, Sharm El Sheikh, Egypt.
  • Sep 2017: Perspectives on Deep Multimodal Robot Learning is accepted as one of the 16 papers selected for ISRR'17 Blue Sky papers.
  • Sep 2017: Topometric Localization with Deep Learning is accepted for ISRR'17.
  • Jun 2017: Our paper on autonomous street crossing for pedestrian assistant robots is accepted for IROS'17.
  • Jan 2017: Successful final demo for the EUROPA2 project. [Video].
  • May 2016: Our work on Probabilistic Camera Localization from Images using Text Spotting is accepted for the Duetsche Gesellschaft fuer Robotik-Tage 2016 (DGR Days 2016).
  • Jan 2016: Our paper on Leveraging Geo-Referenced Texts for Global Localization in Public Maps is accepted for ICRA'16.
Benutzerspezifische Werkzeuge