May 2015 - Present
CHIEF TECHNOLOGY OFFICER & CO-FOUNDER
What started as a casual chat after a game of pick-up soccer, ultimately ended up in a transformational collaboration. Stuart Maggs was introduced to me by my dear friend and soccer teammate James Askew. In 2015, Stuart & I co-founded Naska.AI with a mission to modernize 21st-century construction. We have a very interesting set of products, projects, and open positions. Check out our website for more details.
June 2012 - May 2018
I spent a wonderful 6 years in sunny Southern California working at the (now defunct) Computational Learning and Motor Control Laboratory at the University of Southern California. During this period I was also a visiting researcher at the Max Planck Institute for Intelligent Systems. I also received a Masters' in Computer Science focused on mathematical optimization from USC. My doctoral research primarily focused on algorithms for active and interactive perception for 3D object recognition. You can read my dissertation here.
May 2016 - Aug 2016
I was a part of the computer vision team at Qualcomm Research in San Diego. I worked on algorithms for loop closure detection for visual-inertial navigation. My hosts were Dr. Chris Brunner and Dr. Harris Teague.
May 2015 - Aug 2015
COMPUTER VISION RESEARCHER
At the Mitsubishi Electric Research Labs in Boston, I worked with the computer vision team on Illumination Invariant Image Matching for Autonomous Driving Applications. My hosts were (Prof.) Dr. Srikumar Ramalingam and Dr. Yuichi Taguchi.
Sept 2009 - June 2015
At the University of Pennsylvania, I received my second masters (Robotics). I did research on computer vision and active perception at the world-renowned GRASP Laboratory under the guidance of Prof. Kostas Daniilidis and Prof. Camillo J. Taylor. My thesis title "Sequential Hypothesis Testing for Next Best View Estimation" was supervised by, Prof. Kostas Daniilidis who was also my primary advisor. You can find a copy of my masters' thesis here.
May 2011 - Sept 2011
I spent a summer at the Bosch Research and Technology Center in Palo Alto, working on failure recovery for autonomous mobile robots and humanoids using shared autonomy interfaces. I was hosted by Dr. Benjamin Pitzer and Dr. Sarah Osentoski.
May 2010 - Sept 2010
I worked on Riverine Mapping Project under the advisement of Prof Sanjiv Singh at the Robotics Institute at Carnegie Mellon University. My primary research objective was to detect rivers in monocular color images to enable navigation for UAVs in GPS denied riverine environments. During this period I collaborated with Dr. Stephen Nuske and Dr. Sebastian Scherer.
Aug 2007 - May 2008
I received my first masters in Aerospace Engineering from the University of Maryland, College Park where I worked on Space Robotics and Interplanetary Guidance and Navigation. I also had the pleasure of working at Space Systems Laboratory under the supervision of Dr. David Akin.
Over the years I have had the opportunity to work with some wonderful student interns at Naska.AI, some of whose theses I have advised. My former students and their projects are listed below. If you want to do an internship or a thesis project with me, have a look at the projects I am currently offering.
JOAQUIM ORTIZ DE HARO
PhD Student at IMPRS-IS
Project: Geometric Alignment for Large Scaled Rigid 3D Structures
MARÍA ISABEL ARTIGAS
PhD student at KU Leuven
Project: Robust Arm Pose Estimation for Visual Servoing
Robotics Engineer at Unmanned Life
Masters Thesis (2019)
Project: Robust 3D IMU-LIDAR Calibration and
Multi Sensor Probabilistic State
MARIA VILA ABAD
Software Engineer at StreamSets
Masters Thesis (2020)
Project: Instance Segmentation Annotation on Construction
MONICA PÉREZ SERRANO
Robotics Engineer at Sevensense Robotics
Project: Trajectory Optimization for Remote Assisted Teleoperation
Software Engineer at Naska.AI
Masters Thesis (2021)
Project: Multimodal Semantic SLAM for Lidar Visual Inertial Systems
NAVID KAYHANI, PH.D.
Machine Learning Engineer at Naska.AI
Project: BIM-based construction quality assessment with Graph Neural Networks.