Margarita Grinvald

I am a Staff Computer Vision Engineer on the Physical AI team at Meta, where I work on a 3D vision-language model and on-device spatial AI for next-generation AR/MR experiences.

In 2021, I completed my PhD in Robotics and Computer Vision at ETH Zurich, supervised by Prof. Roland Siegwart.

Email,CV,Google Scholar,GitHub

Publications

Modelify: An approach to incrementally build 3D object models for map completion
Fadri Furrer*, Tonci Novkovic*, Marius Fehr, Margarita Grinvald, Cesar Cadena, Juan Nieto, Roland Siegwart, The International Journal of Robotics Research, 2023.

Paper
Dense object-level robotic mapping
Margarita Grinvald, Doctoral thesis, 2022.

Full text, Slides
TSDF++: A Multi-Object Formulation for Dynamic Object Tracking and Reconstruction
Margarita Grinvald, Federico Tombari, Roland Siegwart, Juan Nieto, International Conference on Robotics and Automation (ICRA), 2021.

Paper, Video, Code
Volumetric Instance-Aware Semantic Mapping and 3D Object Discovery
Margarita Grinvald, Fadri Furrer, Tonci Novkovic, Jen Jen Chung, Cesar Cadena, Roland Siegwart, Juan Nieto, Robotics and Automation Letters, 2019.

★ IROS Best Cognitive Robotics Paper Award Finalist ★

Paper, Video, Slides, Poster, Code
CLUBS - An RGB-D Dataset with Cluttered Box Scenes Containing Household Objects
Tonci Novkovic, Fadri Furrer, Marko Panjek, Margarita Grinvald, Roland Siegwart, Juan Nieto, The International Journal of Robotics Research, 2019.

Paper, Website
Incremental Object Database: Building 3D Models from Multiple Partial Observations
Fadri Furrer, Tonci Novkovic, Marius Fehr, Abel Gawel, Margarita Grinvald, Torsten Sattler, Roland Siegwart, Juan Nieto, International Conference on Intelligent Robots and Systems (IROS), 2018.

Paper, Video

Workshops

Go Fetch: Mobile Manipulation in Unstructured Environments
Kenneth Blomqvist*, Michel Breyer*, Andrei Cramariuc*, Julian Förster*, Margarita Grinvald*, Florian Tschopp*, Jen Jen Chung, Lionel Ott, Juan Nieto, Roland Siegwart, Workshop on Perception, Action, Learning, International Conference on Robotics and Automation (ICRA), 2020.

Paper, Video

Talks

Differentiable Passthrough - A learning-based approach to reduce perceptual artifacts
CVPR 2025 - Workshop on Computer Vision and Mixed Reality, Nashville. June 2025.
Dense Object-Level Mapping
Autonomy Talks, virtual. June 2021.
Dense Object-Level Mapping
NCCR Robotics - Seminar Series on Flying and Legged Robots, virtual. April 2021.
Volumetric Instance-Aware Semantic Mapping and 3D Object Discovery
Amazon Research Awards – Robotics Symposium, Boston. October 2019.

Teaching

Teaching Assistant, ETH Zurich
Perception and Learning for Robotics - Spring 2020
Teaching Assistant, ETH Zurich
Autonomous Mobile Robots - Spring 2019
Teaching Assistant, ETH Zurich
Autonomous Mobile Robots MOOC - Autumn 2018
Teaching Assistant, ETH Zurich
Perception and Learning for Robotics - Spring 2018
Teaching Assistant, ETH Zurich
Autonomous Mobile Robots - Spring 2018
Teaching Assistant, ETH Zurich
Autonomous Mobile Robots MOOC - Autumn 2017

Meetup

Since 2019, I am co-organizer for the Zurich Computer Vision Meetup.

Computer Vision in the age of Robotics
ABB Zurich, July 2024.
Building a digital world with Computer Vision
Meta Zurich, July 2023.
The future through the lenses of Augmented and Virtual Reality
Google Zurich, December 2019.
Learning to Fly - Challenges and results in perception and learning for drones
ETH Zurich, July 2019.