Dr Tobias Fischer

Senior Lecturer (Intelligent robotic systems, Robotic vision, Autonomous vehicles, SLAM)

Doctor of Philosophy (Imperial College, London), M.Sc. (University of Edinburgh), B.Sc. (Other)

Research Overview
Dr Tobias Fischer conducts interdisciplinary research at the intersection of computer vision, cognitive robotics and computational cognition. His research goal is to provide robots with perceptional abilities that allow interactions with humans in a human-like manner. To develop these perceptional abilities, Tobias believes that it is useful to study the principles used by the animal visual system. He uses these principles to develop new computer vision algorithms and validates their effectiveness in intelligent robotic systems.

Research Experience
Dr Fischer is the Principal Investigator of a grant entitled "Spike-based Visual Place Recognition using Intel's Loihi" funded by the Intel Neuromorphic Computing Lab. He is further co-investigator of an Amazon Research Award on "Complementarity-Aware Multi-Process Fusion for Long Term Localization". Dr Fischer was a co-author and named lead researcher of two applications for the Samsung Global Research Outreach program, which resulted in 200,000 USD commercial funding. In addition, he has been working on major research projects funded by European Union FP7 and H2020 programs, and the Multidisciplinary University Research Initiative (MURI). His papers have received two best poster awards:

• Samsung AI Forum 2018 for the paper entitled "Context-aware Deep Feature Compression for High-speed Visual Tracking" (appeared at CVPR2018)
• IEEE International Conference on Computer Vision 2019 Workshop on Gaze Estimation and Prediction in the Wild for the paper entitled "RT-BENE: A Dataset and Baselines for Real-Time Blink Estimation in Natural Environments"

Career History
Before joining QUT as a Research Fellow in January 2020, Dr Fischer was a postdoctoral researcher in the Personal Robotics Lab at Imperial College London. He received a PhD from Imperial College with the thesis topic: "Perspective Taking in Robots: A Framework and Computational Model" in January 2019. The thesis has been awarded the Queen Mary UK Best Thesis in Robotics Award 2018 and the Eryl Cadwaladr Davies Prize for the best thesis in the Electrical and Electronic Engineering Department at Imperial College 2018. Since September 2020, Tobias is a professionally registered Chartered Engineer. Dr Fischer received the M.Sc. degree in Artificial Intelligence from The University of Edinburgh, in August 2014, and a B.Sc. degree in Computer Engineering from Ilmenau University of Technology, Germany, in 2013. He wrote his Bachelor thesis in John Tsotsos' Lab for Active and Attentive Vision, at the York University, Canada. From February 2012 until August 2014, he was a scholarship holder at the prestigious German National Academic Foundation (Studienstiftung des Deutschen Volkes).

Web Links

• Personal website
• Google Scholar profile
• LinkedIn
• Twitter @TobiasRobotics

Publication highlights

1. Hausler, Garg, Xu, Milford & Fischer: Patch-NetVLAD: Multi-Scale Fusion of Locally Global Descriptors for Place Recognition (IEEE Conference on Computer Vision and Pattern Recognition 2021)
2. Fischer & Milford: Event-Based Visual Place Recognition With Ensembles of Temporal Windows (IEEE Robotics and Automation Letters 2020)
3. Fischer & Demiris: Computational Modelling of Embodied Visual Perspective-taking (IEEE Transactions on Cognitive and Developmental Systems)
4. Fischer, Chang & Demiris: RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments (European Conference on Computer Vision ECCV2018)
5. Choi et al.: Context-aware Deep Feature Compression for High-speed Visual Tracking (IEEE Conference on Computer Vision and Pattern Recognition CVPR2018)
6. Chang, Fischer, Petit, Zambelli and Demiris: Learning Kinematic Structure Correspondences Using Multi-Order Similarities (IEEE Transactions on Pattern Analysis and Machine Learning TPAMI2018 & CVPR2016)

Projects (Chief investigator)

• Active, adaptive & predictive robot localization
• Adaptive and Efficient Robot Positioning Through Model and Task Fusion
• Continously updating place descriptors
• Event-based localisation for ultra-fast systems
• Implicit representations for place recognition and robot localisation
• Neuro-Autonomy: Neuroscience-Inspired Perception, Navigation, and Spatial Awareness for Autonomous Robots: AUSMURI project
• Neuromorphic computing
• Neuromorphic sensors and engineering for place recognition
• Re-Evolving Nature’s Best Positioning Systems for People and Their Machines: Australian Research Council Laureate Fellowship 2022-2027

Projects

• AUSMURI: Neuro-Autonomy: Neuroscience-Inspired Perception, Navigation, and Spatial Awareness for Autonomous Robots
• Complementarity-Aware Multi-Process Fusion for Long Term Localization
• NVIDIA Applied Research Accelerator Program
• Re-Evolving Nature’s Best Positioning Systems for People and Their Machines
• Spike-based Visual Place Recognition using Intel’s Loihi
• Weakly Supervised Segmentation of Underwater Imagery

Additional information