(Last updated: October 2020)
You can also find my CV in pdf format here.
2020 - : Senior Manager, Machine Learning Research, Toyota Research Institute (TRI), CA, USA. Head of the Machine Learning Research team at TRI. Applied ML research for Toyota’s future products, including World-scale Autonomous Robots that learn from Experience. Our key research areas include Deep Learning (esp. self-supervised learning), Computer Vision (esp. dynamic 3D scene understanding), and using Simulation for ML (esp. differentiable rendering and sim2real).
2017 - 2019: Manager & Senior Research Scientist, Machine Learning, Toyota Research Institute (TRI), CA, USA. Manager of the ML team at TRI. Responsible for creating and leading TRI’s ML strategy for Automated Driving, from research to ML cloud infrastructure (used in production) and deployment of safety-critical ML models on public roads. Founded and grew the team to a dozen research scientists and ML engineers. Technical advisor for Toyota AI Ventures. External leadership via publications (22 in 2017-2019), invited talks, demonstrations, and scientific collaborations (esp. with Stanford and Preferred Networks).
2013 - 2016: Research Scientist (Computer Vision), Xerox Research Center Europe, Meylan, France. Led research and transfer projects on Deep Learning for video understanding (action recognition, multi-object tracking), domain adaptation, simulation for perception, and Business Process Modeling with Computer Vision.
2008 - 2012: Doctoral Researcher, Microsoft Research - Inria joint center, Paris, France. Invented, implemented, and experimentally validated state-of-the-art Computer Vision and Machine Learning algorithms for action recognition in challenging real-world video sources like movies or YouTube videos. Contributions on how to decompose actions into structured, discriminative, and robust models.
2008: R&D Engineer, LEAR team, Inria, Grenoble, France. Participation to two international Computer Vision competitions: TRECVID and PASCAL VOC (co-winner of classification and detection challenges). Experimentation on tens of thousands of images and videos, using a cluster of computers, under strong time constraints.
06-08/2007: Research intern, Inria Rocquencourt, Paris, France. Research on learning the structure of dynamic Bayesian networks using statistical tests and genetic algorithms.
06-08/2006: Research intern, LIG research lab, IMAG institute, Grenoble, France. Implementation (OCaml) of formal methods in the automatic proof research domain.
- 2008 - 2012: PhD in Computer Science, Microsoft Research - INRIA, Paris \& LEAR Team, INRIA Grenoble, under the supervision of Cordelia Schmid and Zaid Harchaoui, in the fields of Computer Vision and Machine Learning. Title: Structured Models for Action Recognition in Real-world Videos.
- 2007 - 2008: MSc in Artificial Intelligence, Institut Polytechnique (INP), Grenoble, France.
- 2005 - 2008: Engineer Diploma in Computer Science and Applied Mathematics, ENSIMAG (Ecole Nationale Superieure d’Informatique et de Mathematiques Appliquees de Grenoble), France.
- 2003 - 2005: “Classes Preparatoires MPSI et MP*”, preparation courses (Mathematics and Physics) for the French “Grandes Ecoles”, Clermont-Ferrand, France.
- 2003, European Scientific Baccalaureate with distinction, equivalent to “A” levels, Lycee Jeanne d’Arc, Clermont-Ferrand.
- Top 10% reviewer award at NeurIPS 2020
- COCO-Mapillary Competition Runner-Up at ECCV 2018
- Outstanding reviewer award at CVPR 2018
- Outstanding reviewer award at CVPR 2015
- Xerox Innovation Group President’s Award for innovative research in Computer Vision, 2015
- Xerox XTIN grant for high risk - high reward project on “Video Analytics in a Virtual World”, 2014
- Microsoft Research - Inria PhD scholarship grant, 2008 - 2012 2008, Co-winner of the PASCAL VOC 2008 challenge on object classification and detection
Press Coverage and Interviews
- TechXplore article on our IROS’20 paper on A framework to increase the safety of robots operating in crowded environments
- This Week in ML \& AI (TWiMLAI) podcast: “Advancing Autonomous Vehicle Development Using Distributed Deep Learning”, 2019
- SF DL Summit interview on ML at Toyota, 2019
- W\&B interview on ML at TRI, 2019
- AWS Blog: “TRI accelerates safe automated driving with deep learning at a global scale on AWS”, 2018
- Forbes: “Artificial Intelligence: The Clever Ways Video Games Are Used To Train AIs”, 2018
- Forbes: “How Deep Learning Can Use Virtual Worlds To Solve Real World Problems”, 2016
- MIT Tech Review: “To Get Truly Smart, AI Might Need to Play More Video Games”, 2016
- Wired: “Making AI Play Lots of Videogames Could Be Huge (No, Seriously)”, 2016
- El Espanol: “Máquinas más listas gracias a los videojuegos”, 2016
- French (native)
- English (fluent)
- German (used to be fluent, lived in Germany 1999-2002)
- Python (proficient), C/C++, bash
- PyTorch (main DL framework), Docker, Linux, AWS, HPC
- Open Source: contributor to and creator of Open Source Python projects, especially on action recognition, camera motion compensation, and kernel methods (cf. https://github.com/daien and https://github.com/AdrienGaidon-TRI).
- Computer Vision, object detection, tracking, semantic segmentation, action recognition, 3D vision (esp. depth estimation), synthetic data (esp. from simulators and game engines)
- Machine Learning, Deep Learning (esp. convnets), supervised learning, self-supervised learning, domain adaptation, multi-task learning, optimization, kernel methods, time series analysis
- 50+ publications ($1,900+$ citations, cf. below and Google Scholar profile), 19 patents published (8 granted, cf. Google Patents).
- Co-organizer of the ECCV 2020 workshop on Perception for Autonomous Driving (PAD)
- Co-organizer of the ICML 2020 workshop on AI for Autonomous Driving (AIAD)
- Co-organizer of the ICML 2019 workshop on AI for Autonomous Driving (AIAD)
- Guest Editor for the International Journal of Computer Vision (IJCV) Special Issue on “Synthetic Visual Data”
- Co-organizer of the First International Workshop on Virtual/Augmented Reality for Visual Artificial Intelligence (VARVAI) at ECCV 2016 and ACM-MM 2016
- Service: reviewer for the major machine learning, computer vision, and robotics conferences and journals (CVPR, ICLR, ICRA, RSS, NeurIPS, ICML, ICCV, ECCV, BMVC, IJCV, PAMI, TCSVT, …).
- MATS: An Interpretable Trajectory Forecasting Representation for Planning and Control - B. Ivanovic, A. Elhafsi, G. Rosman, A. Gaidon, M. Pavone - CoRL 2020 - November 16, 2020
- Self-Supervised 3D Keypoint Learning for Ego-Motion Estimation - J. Tang, R. Ambrus, V. Guizilini, S. Pillai, H. Kim, P. Jensfelt, A. Gaidon - CoRL 2020 (oral) - November 16, 2020
- Risk-Sensitive Sequential Action Control with Multi-Modal Human Trajectory Forecasting for Safe Crowd-Robot Interaction - H. Nishimura, B. Ivanovic, A. Gaidon, M. Pavone, M. Schwager - IROS 2020 - October 25, 2020
- Game-Theoretic Planning for Risk-Aware Interactive Agents - M. Wang, N. Mehr, A. Gaidon, M. Schwager - IROS 2020 - October 25, 2020
- End-to-end Birds-eye-view Flow Estimation for Autonomous Driving - KH. Lee, M. Kliemann, A. Gaidon, J. Li, C. Fang, S. Pillai, W. Burgard - IROS 2020 - October 25, 2020
- Driving Through Ghosts: Behavioral Cloning with False Positives - A. Bühler, A. Gaidon, A. Cramariuc, R. Ambrus, G. Rosman, W. Burgard - IROS 2020 - October 25, 2020
- Behaviorally Diverse Traffic Simulation via Reinforcement Learning - S. Shiroshita, S. Maruyama, D. Nishiyama, M. Ynocente Castro, K. Hamzaoui, G. Rosman, J. DeCastro, KH. Lee, A. Gaidon - IROS 2020 - October 25, 2020
- Discovering Avoidable Planner Failures of Autonomous Vehicles using Counterfactual Analysis in Behaviorally Diverse Simulation - D. Nishiyama, M. Ynocente Castro, S. Maruyama, S. Shiroshita, K. Hamzaoui, Y. Ouyang, G. Rosman, J. DeCastro, KH. Lee, A. Gaidon - ITSC 2020 - September 20, 2020
- It Is Not the Journey but the Destination: Endpoint Conditioned Trajectory Prediction - K. Mangalam, H. Girase, S. Agarwal, K-H. Lee, E. Adeli, J. Malik, A. Gaidon - ECCV 2020 (oral, top 2%) - August 27, 2020
- Monocular Differentiable Rendering for Self-Supervised 3D Object Detection - D. Beker, H. Kato, MA. Morariu, T. Ando, T. Matsuoka, W. Kehl, A. Gaidon - ECCV 2020 - August 23, 2020
- Neural Ray Surfaces for Self-Supervised Learning of Depth and Ego-motion - I. Vasiljevic, V. Guizilini, R. Ambrus, S. Pillai, W. Burgard, G. Shakhnarovich, A. Gaidon - 3DV 2020 (oral) - August 15, 2020
- Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving - Z. Cao, E. Biyik, W. Z. Wang, A. Raventos, A. Gaidon, G. Rosman, D. Sadigh - RSS 2020 - July 15, 2020
- Differentiable Rendering: A Survey - H. Kato, D. Beker, M. Morariu, T. Ando, T. Matsuoka, W. Kehl, A. Gaidon - arXiv - June 22, 2020
- Spatio-Temporal Graph for Video Captioning with Knowledge Distillation - B. Pan, H. Cai, DA Huang, KH Lee, A. Gaidon, E. Adeli, JC Niebles - CVPR 2020 - June 16, 2020
- Real-Time Panoptic Segmentation from Dense Detections - R. Hou, J. Li, A. Bhargava, A. Raventos, V. Guizilini, C. Fang, J Lynch, A. Gaidon - CVPR 2020 (oral, top 5.7%) - June 16, 2020
- Autolabeling 3D Objects with Differentiable Rendering of SDF Shape Priors” - S. Zakharov, W. Kehl, A. Bhargava, A. Gaidon - CVPR 2020 (oral, top 5.7%) - June 16, 2020
- 3D Packing for Self-Supervised Monocular Depth Estimation - V. Guizilini, R. Ambrus, S. Pillai, A. Raventos, A. Gaidon - CVPR 2020 (oral, top 5.7%) - June 16, 2020
- Spatiotemporal Relationship Reasoning for Pedestrian Intent Prediction - B. Liu, E. Adeli, Z. Cao, KH Lee, A. Shenoi, A. Gaidon, JC Niebles - RA-L & ICRA 2020 - May 31, 2020
- Semantically-Guided Representation Learning for Self-Supervised Monocular Depth - V. Guizilini, R. Hou, J. Li, R. Ambrus, A. Gaidon - ICLR 2020 - April 26, 2020