I’m a PhD student in the Machine Learning Group at the University of Toronto, focusing on computer vision and deep learning for autonomous robotics. I started in September 2017, under the supervision of Professor Raquel Urtasun.

In addition to this, I am also a full-time researcher at Uber Advanced Technologies Group (ATG) Toronto, also led by Professor Urtasun, working on applying my research to the challenges associated with autonomous driving in the real world.

In addition to machine learning and computer vision, my research interests include robotics and long-term autonomy. I am also interested in machine learning security, and I believe that more research is needed in this area (together with its complementary subfield, interpretability), given the growing influence of various machine learning-powered technologies on our daily lives.



  • Exploiting Sparse Semantic HD Maps for Self-Driving Vehicle Localization
    Ioan Andrei Bârsan*

    International Conference on Intelligent Robots and Systems (IROS) 2019
    Note: *denotes equal contribution.
    [PDF] [BibTeX]

    TL;DR: We use very sparse maps consisting in lane graphs (i.e., polylines) and stored traffic sign positions to localize autonomous vehicles. These maps take up ~0.5MiB/km2, compared to, e.g., LiDAR ground intensity images which can take over 150MiB/km2. We use these maps in the context of a histogram filter localizer, and show median lateral accuracy of 0.05m and median longitudinal accuracy of 1.12m on a highway dataset.

  • Learning to Localize through Compressed Binary Maps
    Ioan Andrei Bârsan*

    International Conference on Computer Vision and Pattern Recognition (CVPR) 2019
    Note: *denotes equal contribution.
    [PDF] [BibTeX]

    TL;DR: High-resolution maps, while allowing extremely accurate localization, can also take up a lot of storage. In this paper, we use neural networks to perform task-specific compression to address this issue by learning a special-purpose compression scheme for the specific task of localization. We achieve two orders of magnitude of improvement (0.007 bits/px) over traditional methods like WebP (0.580 bits/px), as well as less than half the bitrate of a general-purpose learning-based compression scheme (0.016 bits/px). For reference, a lossless PNG uses 4.94 bits/px in our dataset.

  • Learning to Localize Using a LiDAR Intensity Map
    Ioan Andrei Bârsan*

    Proceedings of the Second Conference on Robot Learning (CoRL) 2018
    Note: *denotes equal contribution.
    [PDF] [BibTeX]

    TL;DR: Matching-based localization methods using LiDAR can provide centimeter-level accuracy, but require careful beam intensity calibration in order to perform well. In this paper, we cast the matching problem as a learning task and show that it is possible to learn to match online LiDAR observations to a known map without calibrated intensities.

  • Robust Dense Mapping for Large-Scale Dynamic Environments
    Ioan Andrei Bârsan

    Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) 2018
    [Web] [PDF] [Code] [BibTeX]

    TL;DR: A system for outdoor online mapping using a stereo camera capable of also reconstructing the dynamic objects it encounters, in addition to the static map. Supports map pruning to eliminate stereo artifacts and reduce memory consumption.

Work Experience


  • Current: Full-time research scientist at Uber ATG Toronto (Jan 2018–present).
    • Helping develop scalable and robust centimeter-accurate localization methods for self-driving cars.
    • LiDAR-based map localization, visual localization, learning-based compression, large-scale machine learning (Apache Spark).
    • Researching the next generation of autonomous vehicle maps. What can we encode in maps that goes beyond topology and appearance?
  • Previously, I did a series of software engineering internships in the US during my undergrad:
    • Internship: Twitter (Summer 2015, San Francisco, CA), Performance Ads
      • Developed Apache Storm and Hadoop data pipelines using Scala.
    • Internship: Google (Summer 2014, New York, NY), Data Protection
      • Co-developed a system for performing security-oriented static analysis of shell scripts used to run large numbers of cluster jobs.
    • Internship: Microsoft (Summer 2013, Redmond, WA), Server and Tools Business
      • Security and reliability analysis of a web service part of the Azure portal.


  • Teaching Assistant: Image Analysis and Understanding (CSC420), University of Toronto, Fall 2017.
  • Reviewer: ICRA 2019, IROS 2019


Other Projects

  • MetalNet, a small toolkit for scraping and processing metal lyrics, followed by training a language model to generate its own metal. (Source code and blog post coming soon™!)
  • Yeti, an OpenGL 3D game engine with forward and deferred rendering support, real time shadow mapping and more.
  • A bunch of old games I developed for fun can be found on my old Ludum Dare page. It may be tricky to build and run them, though, given the age of the code.


Before starting my PhD, I completed my Master’s in Computer Science at ETH Zurich. For my Master’s Thesis, I developed DynSLAM, a dense mapping system capable of simultaneously reconstructing dynamic and potentially dynamic objects encountered in an environment, in addition to the background map, using just stereo input. More details can be found on the DynSLAM project page.

Previously, while doing my undergraduate studies at Transilvania University, in Brașov, Romania, I interned at Microsoft (2013, Redmond, WA), Google (2014, New York, NY) and Twitter (2015, San Francisco, CA), working on projects related to privacy, data protection, and data pipeline engineering.

I am originally from Brașov, Romania, a lovely little town which I encourage everybody to visit, together with the rest of Southeast Europe.


Email me at iab (at) cs (dawt) toronto (dawt) edu.

Find me on Twitter, GitHub, Google Scholar, LinkedIn, or StackOverflow.