• Changhyun Choi, Ph.D.

• 

Postdoctoral Associate
Computer Science & Artificial Intelligence Lab (CSAIL)
Massachusetts Institute of Technology (MIT)
32 Vassar St, 32-379
Cambridge, MA 02139
USA

PHONE: +1-678-488-1023
EMAIL:
URL: http://people.csail.mit.edu/cchoi
NATIONALITY: Korean
LANGUAGE: Korean and English

Research interests

Visual perception for robotic manipulation, with a focus on 3D object recognition and pose estimation, visual tracking, active perception, visual verification, and 3D registration.

Current Position

• Postdoc, CSAIL, Massachusetts Institute of Technology
• Cambridge, MA, USA

Advisor: Prof. Daniela Rus

Developing a novel visual verification approach for robotic assembly manipulation

Advancing active sensor planning for multi-robot assembly

Combining visual exteroception and proprioception to localize objects in soft hands

Education

• Ph.D. in Robotics, College of Computing, Georgia Institute of Technology
• Atlanta, GA, USA

Thesis: “Visual Object Perception in Unstructured Environments”

Advisor: Prof. Henrik I. Christensen

Thesis Committee: Prof. Dieter Fox, James M. Rehg, Irfan Essa, Anthony Yezzi

GPA: 4.0/4.0

• B.S. in Electrical and Computer Engineering, Sungkyunkwan University
• Suwon, Korea

Thesis: “Real-time 3D Object Pose Estimation and Tracking for Natural Landmark Based Visual Servo” (appeared in IROS'08)

Advisor: Prof. Sukhan Lee

GPA: 4.42/4.5 (Summa cum Laude)

Computer science specialization with an emphasis on computer vision and robotics

Honors & awards

• Graduate Study Abroad Scholarship, Korea Foundation for Advanced Studies (KFAS)
• Seoul, Korea

Full financial support, including tuition and stipend, up to five years for Ph.D. study

• Summa cum Laude, with highest honor in engineering, Sungkyunkwan University
• Suwon, Korea

• 3rd Place in the Small League MiroSot, FIRA Robot Soccer World Championship
• Seoul, Korea

Played as a team representative and was in charge of computer vision and artificial intelligence

• Merit-based Scholarship, Sungkyunkwan University
• Suwon, Korea

Tuition waiver award based on academic excellence for 7 semesters

Research experience

• Robotics and Intelligent Machines (RIM), Georgia Institute of Technology
• Atlanta, GA, USA

Graduate Research Assistant

Advisor: Prof. Henrik I. Christensen

Funding: Boeing Company

Designed and implemented an object pose estimation using 3D point clouds for factory assembly robots

• Google Summer of Code, Google
• Mountain View, CA, USA

Student Developer

Organization: Point Cloud Library (PCL)

Designed and implemented a 3D edge extraction algorithm which detects various types of edges from geometric structure and photometric textures of a given organized scene point cloud

• Imaging Group, Mitsubishi Electric Research Labs (MERL)
• Cambridge, MA, USA

Research Intern

Advisor: Dr. Yuichi Taguchi and Dr. Oncel Tuzel

Designed and implemented a voting-based pose estimation algorithm using a 3D point cloud sensor

• Robotics and Intelligent Machines (RIM), Georgia Institute of Technology
• Atlanta, GA, USA

Graduate Research Assistant

Advisor: Prof. Henrik I. Christensen

Funding: General Motors

Designed and implemented a model-based visual object tracking for the collaborative research with General Motors - Interaction and Learning for Autonomous Assembly Robots

• Imaging Media Research Center, Korea Institute of Science and Technology (KIST)
• Seoul, Korea

Commissioned Research Scientist

Supervisor: Dr. Ig-jae Kim and Dr. Hyoung-gon Kim

Developed a 3D reconstruction software using Internet photos based on Noah Snavely's work

Developed an automatic geotagging software using Internet photos

• Intelligent System Research Center, Sungkyunkwan University
• Suwon, Korea

Undergraduate Research Assistant

Designed and implemented a real-time 3D pose tracking method using KLT and SIFT keypoints

Partially contributed to implementing an object recognition algorithm fusing multiple visual features in a particle filter

• Ubiquitous-VR Lab, Gwangju Institute of Science and Technology
• Gwangju, Korea

Intern Researcher

Developed a hand gesture recognition system using a 3D accelerometer

Employed a support vector machine for gesture classification

Publications

Please refer to PUBLICATIONS tab.

Teaching Experience

• MIT 2.166 - Autonomous Driving, Spring 2016
• Cambridge, MA, USA

Co-instructor

Co-developing course materials covering vision-based control, object detection, localization and mapping, safety and correctness, signaling and coordination, smart infrastructure, and mobility-on-demand

Planning to teach vision-based control and object detection

Invited talks

• GRASP Special Seminar, University of Pennsylvania
• Philadelphia, PA, USA

“Visual Object Perception in Unstructured Environments”

• School of Information & Communication Engineering, Sungkyunkwan University
• Suwon, Korea

“Visual Object Perception in Unstructured Environments”

• School of Mechanical & Aerospace Engineering, Seoul National University
• Seoul, Korea

“Model-based Object Pose Estimation and Tracking using 2D and 3D Visual Information”

• PCL (Point Cloud Library) Tutorial at IROS 2012
• Vila Moura, Portugal

“3D Edge Detection and Registration”

• KORUS Summer School, Sungkyunkwan University
• Suwon, Korea

“Vision for Service Robots”

• International Cognitive Vision Workshop at IROS 2009
• St. Louis, MO, USA

“Cognitive Vision for Efficient Scene Processing and Object Categorization in Highly Cluttered Environments”

Professional service

Reviewer, IEEE Transactions on Robotics (T-RO)

Reviewer, Robotics and Autonomous Systems (RAS) Journal

Reviewer, International Journal of Robotics Research (IJRR)

Reviewer, Image and Vision Computing (IVC) Jornal

Reviewer, IEEE International Conference on Robotics and Automation (ICRA)

Reviewer, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Tracking Chair, International Symposium on Mixed and Augmented Reality (ISMAR)