Our goal is to enable Embodied AI and robot learning inspired by the human learning process. The current work focuses on foundation models for building efficient scene representations and simulation environments as well as robot learning in these environments. We are also interested in novel perception and sensing modalities. Overall, our research spans basic problems in machine learning, computer vision, and robotics. Current interests involve the following topics:

• Robotics: Large-scale Models for Robotic Perception and robust Policy Learning.
• Computer Vision: Efficient & Editable Neural 3D Reconstruction, Generative Scene Representations, Neuromorphic Vision.
• Machine Learning: Geometric Deep Learning, Causal Representation Learning, Reinforcement Learning, Imitation Learning.

• Sep 1, 2024, Kai He, Shuhong Zheng, Justin Wu, Jasper Gerigk, and Alfred Gu joined our lab as graduate students.
• Aug 12, 2024, Our paper Watch Your Steps was selected for an Oral presentation at ECCV.
• May 14, 2024, Our papers Watch Your Steps and on Accelerating Online Mapping and Behavior Prediction were accepted at ECCV.
• May 29, 2024, Our work on Map Uncertainty in Trajectory Prediction was selected as a Best Paper Finalist at CVPR.
• May 14, 2024, Our papers on Auxillary Tasks and High Update Ratios in Reinforcement Learning were accepted at RLC.
• May 14, 2024, Vid2Robot was accepted at RSS.
• Feb 26, 2024, Our papers on Multiview Diffusers, Map Uncertainty in Trajectory Prediction, and Label-Efficient Object Detection for Event Cameras were accepted at CVPR.
• Jan 01, 2024, Evgenii Oprysho joined our lab as graduate student.
• Dec 20, 2024, Jasper Gerigk received an Honorable Mention for the CRA Outstanding Undergraduate Researcher Award.
• Dec 8, 2023, The code for GeoMatch is now released here.
• Oct 24, 2023, AvatarOne accepted at WACV.
• Oct 1, 2023, iNVS accepted at SIGGRAPH Asia Conference Track.
• Sep 22, 2023, SlotDiffusion accepted as a Spotlight at NeurIPS and trajdata accepted in the datasets and benchmarks track.
• Sep 01, 2023, Umangi Jain joined our lab as graduate student.
• Aug 30, 2023, Our papers on Multi-team Racing and Geometry Matching for Multi-Embodiment Grasping have been accepted at CoRL.
• Jul 14, 2023, Our paper on Reference-guided controllable inpainting has been accepted at ICCV.
• Jun 04, 2023, Talk at IV Symposium’s Workshop on Socially Interactive Autonomous Mobility.
• Jun 01, 2023, Our workshop paper “Towards Zero-Shot Multi-Team Racing: Competitive Driving via Learning in Simulation” has won the Best Workshop Paper Award at the ICRA Workshop on Multi-Robot Learning.
• Apr 23, 2023, An extended version of our RSS'21 work GROUNDED has been accepted at IJRR.
• Apr 08, 2023, Multi-Abstractive Neural Controller accepted at RA-L.
• Feb 27, 2023, SPIn-NeRF, INS, and SparsePose accepted at CVPR.
• Jan 20, 2023, Solving Continuous Control via Q Learning accepted at ICLR.
• Jan 17, 2023, Talk at Technion Robotics Seminar.
• Jan 07, 2023, We are hosting the Woodlands Secondary School’s FIRST Robotics Competition team “Absolute Robotics” for their season kick-off at UTM.
• Jan 02, 2023, We are offering courses on Robotic Perception and Causal Representatiaon Learning in Winter 2023.
• Jan 01, 2023, Claas Voelcker joined our lab as graduate student.
• Oct 28, 2022, Talk at UZH Robotics & Perception Group.
• Sep 13, 2022, Talk at ETH Autonomy Talks (Video) on “Interactive Navigation for Robust Autonomy”.
• Sep 01, 2022, Andrei Ivanovic, Samarth Sinha, and Ziyi Wu joined our lab as graduate students.
• Jul 08, 2022, LaTeRF and Housekeep accepted at ECCV.
• Jun 30, 2022, Maplite 2.0 accepted at IROS and RA-L.
• Jun 5, 2022, Talk at IV2022 Workshop on Bridging the gap between map-based and map-less driving
• May 26, 2022, Prof. Gilitschenski has been recognized as “Outstanding Associate Editor” for the Robotics and Automation Letters
• Apr 6, 2022, Talk at NVIDIA AV Research Group
• Mar 1, 2022, Neighborhood Mixup Experience Replay accepted at L4DC
• Feb 18, 2022, Talk at Samsung AI Research Toronto
• Jan 31, 2022, Five papers accepted at ICRA: VISTA 2.0, Interactive Driving Policies, HYPER, Concept Graph Network, and AGN (which has already been accepted to RA-L).
• Nov 22, 2021, Our paper on Automaton Generative Networks has been accepted at RA-L.
• Sep 1, 2021, Start of our lab with the first cohort of graduate students: Maria Attarian, Yash Kant, and Ashkan Mirzaei.