My research interests include a variety of topics including both computer vision and human vision, visual attention, machine learning and computational neuroscience. I am interested in applying this knowledge to robotics, to imitate human-like behaviour.
– info (at) tobiasfischer (dot) info
– t (dot) fischer (at) imperial (dot) ac (dot) uk
News
- Nov 2018 – ROS code for our TRACA visual object tracking paper is now available on Jongwon Choi’s website.
- Sept 2018 – Amazing news: our TRACA paper won the best poster award at the Samsung AI Forum! Very proud of the main author Jongwon Choi.
- Sept 2018 – I had a great time at ECCV2018, and it was nice meeting many new people and having stimulating conversations. Also very happy that our work on gaze estimation received a lot of interest!
- Sept 2018 – With only one month before submission of my PhD thesis, I am happy to announce that I will continue my research in the Personal Robotics Lab for some more time.
- Jun 2018 – Great news with a third paper accepted within a single month! This one is at ECCV2018. Tobias Fischer, Hyung Jin Chang and Yiannis Demiris: “RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments”. Download or watch the video. The dataset and code will be released within the next weeks.
- Jun 2018 – Delighted that our IROS2018 submission got accepted! Phuong Nguyen, Tobias Fischer, Hyung Jin Chang, Ugo Pattacini, Yiannis Demiris, and Giorgio Metta: “Transferring Visuomotor Learning from Simulation to the Real World for Robotics Manipulation Tasks”. Download or watch the video .
- Jun 2018 – A workshop paper got accepted to the CVPR2018 Workshop on Vision Meets Cognition. Tobias Fischer and Yiannis Demiris: “A Computational Model for Embodied Visual Perspective Taking: From Physical Movements to Mental Simulation”. Join on Monday 18 June for my spotlight presentation and the poster session! Download paper or the poster
- Feb 2018 – Our CVPR 2018 paper got accepted. Jongwon Choi, Hyung Jin Chang, Tobias Fischer, Sangdoo Yun, Kyuewang Lee, Jiyeoup Jeong, Yiannis Demiris, Jin Young Choi: “Context-aware Deep Feature Compression for High-speed Visual Tracking”. Download or visit project page with (external link)
- Nov 2017 – Our paper entitled “Learning Kinematic Structure Correspondences Using Multi-Order Similarities” got accepted in the IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). Download