photo


I am currently an ML Research Scientist and Tech. Lead at Toyota Research Institute. I received my PhD in CS from MIT where I worked with John Leonard on SLAM-aware, self-supervised perception and learning in mobile robots. Prior to coming to MIT, I was a computer vision developer at PhaseSpace Motion Capture for 2 years working on real-time computer-vision related technologies. I completed my B.S. in Mechanical Engineering in ‘08 at the University of Michigan - Ann Arbor. I’ve also had the opportunity to work as a research intern at some exciting companies like Mitsubishi Electric Research Labs (MERL) and Segway.

I am generally interested in developing self-supervised learning algorithms in SLAM-aware robots, in order to learn persistently in an environment from visual experience. I am interested in high-impact ventures leveraging computer vision, machine learning and perception technologies in autonomous systems (self-driving cars, house-hold robotics, UAVs, AUVs etc). For a more comprehensive overview of my background and previous work, please refer to my CV.

Present 2011 - 2017 2014 2009 - 2011 2007 2005 - 2008

Publications

project-thumb

SuperDepth: Self-Supervised, Super-Resolved Monocular Depth Estimation
S. Pillai, R. Ambrus, and A. Gaidon
International Conference on Robotics and Automation (ICRA), 2019
[ pdf, video ]

project-thumb

SLAM-Aware, Self-Supervised Perception in Mobile Robots
S. Pillai
PhD Thesis, September 2017
[ pdf, slides ]

project-thumb

Self-Supervised Place Recognition in Mobile Robots
S. Pillai and J. Leonard
Learning for Localization and Mapping Workshop, IROS, 2017
[ pdf ]

project-thumb

Towards Visual Ego-motion Learning in Robots
S. Pillai and J. Leonard
International Conference on Intelligent Robots and Systems (IROS), 2017
[ pdf ]

project-thumb

Towards Richer, and Self-Supervised Perception in Robots
S. Pillai
Ph.D Thesis Proposal, Apr 2017
[ pdf ]

project-thumb

SLAMinDB: Centralized graph databases for mobile robotics
D. Fourie, S. Claassens, S. Pillai, R. Mata, J. Leonard
International Conference on Robotics and Automation (ICRA), 2017
[ pdf ]

project-thumb

High-Performance and Tunable Stereo Reconstruction
S. Pillai, S. Ramalingam and J. Leonard
International Conference on Robotics and Automation (ICRA), 2016
[ pdf, video, bib ]

project-thumb

Monocular SLAM Supported Object Recognition
S. Pillai and J. Leonard
Robotics: Science and Systems (RSS), 2015
[ pdf, bib, video, slides, poster ]

project-thumb

Line-Sweep: Cross-Ratio for Wide-Baseline Matching and 3D Reconstruction
S. Ramalingam, M. Antunes, D. Snow, G. H. Lee, S. Pillai
Computer Vision and Pattern Recognition (CVPR), 2015
[ pdf, bib, video ]

project-thumb

Towards Enabling a Robot to Effectively
Assist People in Human-Occupied Envioronments

S. Hemachandra, R. Finman, S. Pillai, M. Walter, S. Teller
AAAI Spring Symposium: Designing Intelligent Robots, March 2012
[ pdf, bib ]