Raquel Urtasun
Toyota Technological Institute at Chicago, University of Toronto
Papers
20
Total Citations
10,349
H-Index
11
About
Raquel Urtasun is a world-renowned leader in computer vision and machine learning for autonomous driving. As a Professor at the University of Toronto and the founder of the Toronto AI Lab at Uber ATG, her research focuses on enabling self-driving vehicles to perceive and understand the world in 3D. Her seminal contribution, the KITTI dataset (2013), with nearly 10,000 citations, revolutionized the field by providing a standardized benchmark for mobile robotics and autonomous driving, becoming the gold standard for evaluating perception algorithms. Urtasun’s work consistently pushes the boundaries of 3D scene understanding, from developing generative models for urban scenes to pioneering deep learning methods for scene flow estimation and 3D object detection from stereo cameras. Her recent innovations, such as LaneRCNN for graph-centric motion forecasting and NeuSim for realistic sensor simulation, address critical challenges in predicting actor behavior and generating high-fidelity virtual environments for testing. With over 10,000 total citations and numerous best paper awards, Urtasun’s research bridges the gap between academic rigor and real-world deployment, making her a pivotal figure in the quest for safe, scalable autonomous driving.
Research Focus
Key Achievements
Top Papers
- 1Vision meets robotics: The KITTI dataset9,681 citations · 2013
- 2LaneRCNN: Distributed Representations for Graph-Centric Motion Forecasting200 citations · 2021
- 3Deep Rigid Instance Scene Flow133 citations · 2019
- 4
- 5A generative model for 3D urban scene understanding from movable platforms57 citations · 2011
- 6PLUMENet: Efficient 3D Object Detection from Stereo Images38 citations · 2021
- 7Asynchronous Multi-View SLAM28 citations · 2021
- 8Learning to Recognize Objects from Unseen Modalities25 citations · 2010
- 9Mending Neural Implicit Modeling for 3D Vehicle Reconstruction in the Wild24 citations · 2022
- 10Reconstructing Objects in-the-wild for Realistic Sensor Simulation17 citations · 2023