|
Associate Research Professor in the Robotics Institute,
School of Computer Science;
Courtesy Professor in the Electrical and Computer Engineering (ECE) department; Director of the Master of Science Computer Vision (MSCV) Program Carnegie Mellon University Research Fellow in the Institute of Industrial Science at the University of Tokyo Research Areas: Computer Vision, Machine Learning, Autonomous Systems |
 |
 |
 |
 |
The vision of our lab is to realize robust autonomous systems built for real-world perception and interactive decision-making. The key focus areas of our lab are perception, decision-making and interaction. Our focus is broad because we believe that innovating at the system-level requires expertise and integration at the component level.
We innovate across the full spectrum of perception including vision-based human pose estimation, action recognition, object detection/tracking/forecasting, 3D scene understanding. We develop computational models for reinforcement learning, inverse reinforcement learning, imitation learning, game-theoretic modeling and Neural Architecture Search. We build real-world systems to enable cyber-physical interaction including wearable camera systems, multi-modal sensors, portable navigational aides and assistive mobile robots.
Hiring Information:
(May 2020) CMU students: I will be hiring at all levels in the Fall. Please contact me during orientation week to setup a brief intervew.
(May 2020) Postdocs: Looking for researchers with experience in multi-agent forecasting, inverse reinforcement learning (structural parameter estimation) or simulated humanoid control.
News
Mar. 2020 3 Papers Accepted to CVPR 2020!
Feb. 2020 Paper on Multi-Agent Multi-group Imitation Learning Accepted to AAMAS 2020!
Jan. 2020 Paper on Diverse Trajectory Forecasting Accepted to ICCV 2019!
Dec. 2019 Two Papers Accepted to CHI 2020!
Aug. 2019 3 Papers Accepted to ICCV 2019!
Feb. 2019 Paper on Network Compression using BO accepted to ICLR 2019!
Research Projects
|
Online Inverse Reinforcement Learning for First-Person Vision |
|
Navigation for the Blind |
|
First-person Vision Activity Recognition |
|
Learning Actions Maps |
|
Game Theoretic Multi-Agent Forecasting |
|
Forecasting Wide Receiver Trajectories |
|
Pedestrian Detection by Synthesis |
|
Visual Grasp Analysis |
|
EgoAction Analysis |
|
BallCam! |
|
Activity Forecasting |
|
Forecasting Interactions |
|
Hand Detection and Analysis |
Research Sponsors
Shimizu Corporation
DENSO Corporation
Caterpillar Inc.
Facebook Reality Labs
Amazon
Highmark / Allegheny General Hospital
National Science Foundation (NSF) - NRI
Japan Science and Tecnology Agency (JST) - CREST, AIP
Intelligence Advanced Research Projects Activity (IARPA) - DIVA
Mational Instite of Disability, Independent Living and Rehabilitation Research (NIDILRR)
Department of Homeland Security (DHS)