Matthew Beveridge

PhD Student in Computational Imaging
Columbia University

beveridge [at] cs.columbia.edu

About

I am a second year PhD student at Columbia University in the Columbia Imaging and Vision Lab (CAVE) with Shree Nayar. My research is in computational imaging, with a focus on understanding the material properties of our lived environment. I'm interested in its relevance to robotics and remote sensing.

Prior to starting at Columbia, I completed my BS and MEng at MIT working with Daniela Rus on robot vision. I also worked on depth estimation at NODAR for a bit, and spent a summer at NASA.

News

[May 2025] We released Matador: A Material Image Dataset. [webpage]

[Jul 2023] I was awarded the Greenwoods Fellowship to support my studies at Columbia.

[Jun 2023] I joined Morteza Karimzadeh as a Visiting Researcher at CU Boulder to work on Arctic sea ice mapping from satellite imagery.

[May 2023] I received the LEAP Momentum Fellowship to study the optical properties of atmospheric aerosols. I will be working with Carl Vondrick and Kara Lamb. [press]

[Apr 2023] This Fall, I will be attending Columbia University to pursue a PhD with Shree Nayar.

[Oct 2022] A proposal for explainable forecasting of Arctic sea ice concentration was accepted to the NeurIPS 2022 Workshop on Tackling Climate Change with Machine Learning. [webpage]

[Sep 2022] Our work on motion tracking to identify and track turbulent filaments in fusion plasmas was accepted to Scientific Reports. [webpage] [press]

[Jan 2022] A paper on bayesian optimization of droplet generating hardware at varying scales was accepted to ACS Applied Materials and Interfaces. [webpage]

[Dec 2021] Two new papers accepted to the NeurIPS 2021 Workshop on Tackling Climate Change with Machine Learning on the topics of the Atlantic Multidecadal Variability [webpage] and biogeochemistry of the Southern Ocean [webpage]. The former was awarded best paper.

[Nov 2021] Generalizing Imaging through Scattering Media with Uncertainty Estimates was accepted to the WACV 2022 Workshop on Applications of Computational Imaging. [pdf]

[Aug 2021] Our work on using generative models to speed up computational fluid dynamics simulation, with application to urban wind modeling, was accepted to ACML 2021 for a long oral spotlight. [webpage]

[Jul 2021] Our paper on image-based audio augmentation was accepted to ICCV 2021. [webpage] I also began work as a Computer Vision Engineer at NODAR.

[May 2021] I completed my Master's thesis on consistent monocular depth estimation in data-driven simulation for reinforcement learning-based autonomous driving. [webpage]

Research

Recently, I have been interested in how to determine the material of an object using optical sensing. Prior to this, most of my work was in the realm of vision for autonomous driving and geospatial machine learning.

javascript disabled, raw bibfile below:

Last updated: May 29^th, 2025.