I am currently at Zoox Inc working in the Calibration, Localization and Mapping team.
Previously I have completed my PhD in Computer Science at University of Maryland, College Park. I was a member of the Computer Vision Lab working with Yiannis Aloimonos and Cornelia Fermüller.
I got my undegraduate degree from the Electrical Engineering Department at Imperial College London, working with Yiannis Demiris.
My research interests lie at the intersection of the fields of Computer Vision and Robotics. In particular I am interested in the problems of object segmentation, recognition, tracking and manipulation by an active robot. In the past I have worked on the problem of shadow free segmentation in still images.
CV | Github
Contact: aleksandrs ecins AT gmail dot com
News and Announcements
| June 2018: | Check-out Cilnatro, a minimal and efficient C++ library for pointcloud mainpulation and visualization. Developed by Konstantinos Zampogiannis at UMD. |
| Nov 2017: | We have participated in the Detecting Symmetry in the Wild challenge held as part of ICCV 2017. Our entry has recieved first place in all 3D symmetry detection categories! Check out the results and the paper detailing our approach [PDF]. |
| Oct 2017: | I have joined Zoox Inc as a research engineer in the CLAMS team (Calibration Localization and Mapping Simultaneoulsy). |
| Jul 2017: | I have succesfully defended my thesis! It is available for download here: Seeing Behing The Scene: Using Symmetry To Reason About Objects In Cluttered Environments. |
Publications
-
Seeing Behing the Scene: Using Symmetry To Reason About Objects in Cluterred Environments
Aleksandrs Ecins, Cornelia Fermüller, Yiannis Aloimonos.
International Conference on Intelligent Robots (IROS), October 2018.
[PDF] [Project page] [Code] -
Detecting Reflectional Symmetries in 3D Data Through Symmetrical Fitting
Aleksandrs Ecins, Cornelia Fermüller, Yiannis Aloimonos.
Detecting Symmetry in the Wild Workshop (ICCV), October 2017.
First place in 3D symmetry detection challenge for real and synthetic 3D data categories!
[PDF] [ICCV challenge results] -
Cluttered Scene Segmentation Using the Symmetry Constraint
Aleksandrs Ecins, Cornelia Fermüller, Yiannis Aloimonos.
International Conference on Robotics and Automation (ICRA), May 2016.
[PDF] [Project page] -
Shadow Free Segmentation in Still Images Using Local Density Measure
Aleksandrs Ecins, Cornelia Fermüller, Yiannis Aloimonos.
International Conference on Computational Photography (ICCP), May 2014.
[PDF] [Project page] [Code] -
Shadow Free Segmentation in Still Images Using Local Density Measure
Aleksandrs Ecins, Cornelia Fermüller, Yiannis Aloimonos.
Perceptual Organization Workshsop (CVPR), June 2014.
[PDF] [Project page] -
Towards a Robotic Hand Rehabilitation Exoskeleton for Stroke Therapy
Yeongjin Kim, Shing Shin Cheng, Aleksandrs Ecins, Cornelia Fermüller, Kelly P Westlake, Jaydev P Desai.
ASME Dynamic Systems and Control Conference, October 2014.
[PDF]
Thesis
-
Seeing Behing The Scene: Using Symmetry To Reason About Objects In Cluttered Environments
Aleksandrs Ecins
Doctoral Thesis, University of Maryland, August 2017
[Project Page][PDF]
Teaching
- Robotics and Perception (CMSC498) under Dr. Cornelia Fermüller, Spring'16 at UMD
- Computer Processing of Pictorial Information (CMSC733) under Prof. Yiannis Aloimonos, Fall'12 at UMD
- Machine Learning (CMSC726) under Prof. Hal Daumé III, Fall'11 at UMD
- Computational Linguistics I (CMSC723) under Dr. Kristy Hollingshead, Fall'11 at UMD
- Object Oriented Programming I (CMSC131) under Prof. David Jacobs, Spring'11 at UMD
- Discrete Structures (CMSC250) under Prof. Don Perlis, Fall'10 at UMD