Profile picture

Dr Tobias Fischer

Lecturer (Assistant Professor)
Queensland University of Technology
I am passionate about my overarching research goal: providing robots with perceptual abilities that allow safe, intelligent interactions with humans in real-world environments. To develop these perceptual abilities, I believe it is useful to study the principles of the human, animal and insect visual systems. I use these principles to develop new computer vision and machine learning algorithms and validate their effectiveness in intelligent robotic systems. I am enthusiastic about this forward/reverse engineering approach, which combines concepts from computer science and engineering with those from biology and social sciences, as it offers the dual benefit of uncovering principles inherent in the animal visual system, as well as applying these principles to its artificial counterpart. I am a recipient of two best thesis awards and two best poster awards and have acquired grants in excess of 400,000 AUD.
Contact Me:  
Tobias Fischer avatarTobias Fischer@TobiasRobotics
In a range of experiments on four challenging datasets, we show that our spiking net achieves comparable performance to the #SoTA, and degrades gracefully in performance with an increasing number of reference places. 4/n
Tobias Fischer avatarTobias Fischer@TobiasRobotics
But place recognition is a harder problem with many more classes. To deal with this, we introduce a novel weighted assignment scheme using ambiguity-informed #salience: Does a #neuron represent (i.e. fire for) just a single place or multiple places? 3/n
Tobias Fischer avatarTobias Fischer@TobiasRobotics
We considered #placerecognition as a #classification task: Which of N places is the robot in, given significant appearance change since last observing this place? Classification has long been studied with spiking nets, e.g. in digit recognition. 2/n
Tobias Fischer avatarTobias Fischer@TobiasRobotics
First work trip since arriving @QUT in @brisbanecityqld. Excited like a small child! Immediate throwbacks seeing #finalcall signs before boarding even started. Many thanks to @Exaptec for inviting me to speak at the #Generate22 #robotics & #AI conference in @Melbourne
Tobias Fischer avatarTobias Fischer@TobiasRobotics
Here we go: @rosorg #ROS2 on a new shiny @Apple M1 Silicon (osx-arm64) thanks to @RoboStack and @condaforge. Still not quite ready for the public (some issues in the installation process), but we are nearly there! Stay tuned and follow @RoboStack to get updates.
Tobias Fischer avatarTobias Fischer@TobiasRobotics
Incredibly happy that our @RoboStack paper has been accepted to the @ieeeras Robotics & Automation Magazine 🥳. @RoboStack brings together #ROS @rosorg with @condaforge and @ProjectJupyter. Preprint: arxiv.org/abs/2104.12910. Find out some key benefits in this 🧵: 1/n
Tobias Fischer avatarTobias Fischer@TobiasRobotics
New research out on #spiking #neuralnetworks (SNNs) for place recognition: arxiv.org/abs/2109.06452 - by @Somayeh_HS with @maththrills. SNNs offer compelling advantages, including #energyefficiency and low #latencies, inspired by how our #brain processes information. 1/2
Tobias Fischer avatarTobias Fischer@TobiasRobotics
Excited to share our new work on "Predicting Secondary Task Performance for Cognitive Overload Detection" - Early access: ieeexplore.ieee.org/document/95429… Led by @PVAmadori we explored task performance in a dual-task decision-making driving scenario, aiming for safer shared-control. 1/n
Tobias Fischer avatarTobias Fischer@TobiasRobotics
Very excited to be selected as a participant in this year's Fresh Science workshop. We'll learn about #science #communication in two full-day events with media training, meet&greet with businesses and a presentation at a local pub. Looking forward to meet the other participants! twitter.com/EconnectTeam/s…
Tobias Fischer avatarTobias Fischer@TobiasRobotics
It's been a busy few days for our brilliant PhD student @Somayeh_HS: ICRA+RA-L deadline last Friday, and tomorrow she will present at the @QUT #3MT finals! Keen to hear about spiking neural nets for place recognition? Register: eventbrite.com.au/e/qut-graduate… @QUTRobotics @maththrills