Niko Sünderhauf

I am a Chief Investigator of the Australian Centre for Robotic Vision, and a Senior Lecturer (roughly equivalent to a tenured Assistant Professor position in the US system) at Queensland University of Technology (QUT) in Brisbane, Australia.

I conduct research in robotic vision, at the intersection of robotics, computer vision, and machine learning. My research interests focus on scene understanding, semantic SLAM, and new ways to incorporate semantics into reinforcement learning. I furthermore lead a project on new benchmarking challenges in robotic vision.

I regularly organise workshops at robotics and computer vision conferences, such as RSS, CVPR, and ICRA. I have been Associate Editor for the IEEE Robotics and Automation Letters journal (RA-L) since 2015, and served as AE for the IEEE International Conference on Robotics and Automation (ICRA) in 2018.

news

Feb 10, 2019

I will be involved in organising two workshops at CVPR in Long Beach this year: Our Robotic Vision Probabilistic Object Detection Challenge, and a workshop on Deep Learning for Semantic Visual Navigation.

Feb 1, 2019

Our papers Evaluating Merging Strategies for Sampling-based Uncertainty Techniques in Object Detection (led by Dimity Miller) and Look No Deeper: Recognizing Places from Opposing Viewpoints under Varying Scene Appearance using Single-View Depth Estimation (led by Sourav Garg) have been accepted for publication at the ICRA 2019.

Sep 2, 2018

CoRL accepted our paper Learning Deployable Navigation Policies at Kilometer Scale from a Single Traversal. This is joint work led by Jake Bruce, in collaboration with Piotr Mirowski and Raia Hadsell from DeepMind, and Michael Milford from QUT.

Aug 20, 2018

Our paper on QuadricSLAM: Dual Quadrics from Object Detections as Landmarks in Object-oriented SLAM has been accepted for publication in the IEEE Robotics and Automation Letters (RA-L) journal. You can read the preprint on arXiv or the accepted version at IEEE. This work is led by Lachlan Nicholson.

Jun 30, 2018

We successfully ran a workshop on New Benchmarks, Metrics, and Competitions for Robotic Learning at RSS in Pittsburgh, and one workshop on Real-World Challenges and New Benchmarks for Deep Learning in Robotic Vision at CVPR in Salt Lake City.