Dominic Roberts

I am an Applied Scientist at Amazon, where I work on the Just Walk Out (JWO) technology used in Amazon Go and Fresh stores.

I obtained my PhD from UIUC where I worked with Mani Golparvar-Fard and David Forsyth on computer vision and its application to automated activity analysis of construction resources.

Before that, I did an MSc in Data Science and Artificial Intelligence at École Centrale de Lille and an MSc in Applied Mathematics at Université Lille 1, both in Lille, France.

Email  /  CV  /  Google Scholar

profile photo
Affiliations
                       
Amazon
2021 - present
Autodesk AI Lab
Summer 2020
AutonomouStuff
Summer 2017
UIUC
2016-2021
École Centrale de Lille
2011 - 2015
Université Lille 1
2014 - 2015
Selected publications
LSD-StructureNet: Modeling Levels of Structural Detail in 3D Part Hierarchies
Dominic Roberts, Ara Danielyan, Hang Chu, Mani Golparvar-Fard, David Forsyth
ICCV 2021

An augmented version of StructureNet that can re-generate parts situated at arbitrary positions in the hierarchies of its outputs.

Vision-based construction worker activity analysis informed by body posture
Dominic Roberts, Wilfredo Torres Calderon, Shuai Tang, Mani Golparvar-Fard
Journal of Computing in Civil Engineering, 2020

A vision-based activity analysis method that leverages the 2D pose estimation outputs used in many state-of-the-art construction worker ergonomics analysis methods, resulting in improved performance.

End-to-end vision-based detection, tracking and activity analysis of earthmoving equipment filmed at ground level
Dominic Roberts, Mani Golparvar-Fard
Automation in Construction, 2019

A framework performing object detection, object tracking and action segmentation to automatically localize and identify earthmoving equipment and the activity they are performing in each video frame.

Annotating 2D imagery with 3D kinematically configurable assets of construction equipment for training pose-informed activity analysis and safety monitoring algorithms
Dominic Roberts, Yunpeng Wang, Ali Sabet, Mani Golparvar-Fard
ASCE International Conference on Computing in Civil Engineering (i3CE), 2019

A prototype of an annotation tool allowing users to annotate real-world images of construction equipment with semantic segmentation masks and keypoints, given a 3D virtual model of the equipment.

An Annotation Tool for Benchmarking Methods for Automated Construction Worker Pose Estimation and Activity Analysis
Dominic Roberts, Mingzhu Wang, Wilfredo Torres Calderon, Mani Golparvar-Fard
International Conference on Smart Infrastructure and Construction (ICSIC), 2019

A 2D human pose annotation tool adapted from CVAT that can also annotate per-frame activity labels.



Credit to Jon Barron and Unnat Jain for this website's template.