Thomas J. Cashman

Researcher, Graphcore

Joined

July 2019

Names

Thomas J. Cashman

Tom Cashman

Thomas Joseph Cashman

Emails

****@microsoft.com (Confirmed)

****@cantab.net (Confirmed)

****@graphcore.ai (Confirmed)

Personal Links

Career & Education History

Researcher

Graphcore (graphcore.ai)

2025 – 2026

Scientist

Microsoft (microsoft.com)

2015 – 2025

Software engineer

TranscenData Europe Ltd (transcendata.com)

2012 – 2015

Postdoc

Faculty of Informatics, Università della Svizzera Italiana (usi.ch)

2010 – 2012

PhD student

Computer Laboratory, University of Cambridge (cam.ac.uk)

2006 – 2010

Advisors, Relations & Conflicts

Postdoc Advisor

Kai Hormann

2010 – 2012

PhD Advisor

Neil A. Dodgson

2006 – 2010

Expertise

3D vision

model-fitting

optimization

Present

human understanding

human-centric vision

skeletal tracking

Present

monocular depth estimation

single-view reconstruction

Present

computer graphics

geometry

subdivision surfaces

Present

Publications

FLAG: Flow-based 3D Avatar Generation from Sparse Observations
Sadegh Aliakbarian, Pashmina Cameron, Federica Bogo, Andrew W. Fitzgibbon, Thomas J. Cashman
- 2022 (modified: 03 Nov 2022)
- CVPR 2022
- Readers: Everyone
3D Face Reconstruction with Dense Landmarks
Erroll Wood, Tadas Baltrusaitis, Charlie Hewitt, Matthew Johnson, Jingjing Shen, Nikola Milosavljevic, Daniel Wilde, Stephan J. Garbin, Toby Sharp, Ivan Stojiljkovic, Tom Cashman, Julien P. C. Valentin
- 2022 (modified: 11 Nov 2022)
- ECCV (13) 2022
- Readers: Everyone
Full-Body Motion from a Single Head-Mounted Device: Generating SMPL Poses from Partial Observations
Andrea Dittadi, Sebastian Dziadzio, Darren Cosker, Ben Lundell, Thomas J. Cashman, Jamie Shotton
- 2021 (modified: 12 Nov 2022)
- ICCV 2021
- Readers: Everyone
Fake it till you make it: face analysis in the wild using synthetic data alone
Erroll Wood, Tadas Baltrusaitis, Charlie Hewitt, Sebastian Dziadzio, Thomas J. Cashman, Jamie Shotton
- 2021 (modified: 27 Dec 2022)
- ICCV 2021
- Readers: Everyone
HoloLens 2 Research Mode as a Tool for Computer Vision Research
Dorin Ungureanu, Federica Bogo, Silvano Galliani, Pooja Sama, Xin Duan, Casey Meekhof, Jan Stühmer, Thomas J. Cashman, Bugra Tekin, Johannes L. Schönberger, Pawel Olszta, Marc Pollefeys
- 2020 (modified: 11 Nov 2022)
- CoRR 2020
- Readers: Everyone
A high fidelity synthetic face framework for computer vision
Tadas Baltrusaitis, Erroll Wood, Virginia Estellers, Charlie Hewitt, Sebastian Dziadzio, Marek Kowalski, Matthew Johnson, Thomas J. Cashman, Jamie Shotton
- 2020 (modified: 27 Dec 2022)
- CoRR 2020
- Readers: Everyone
An Efficient Background Term for 3D Reconstruction and Tracking with Smooth Surface Models
Mariano Jaimez, Thomas J. Cashman, Andrew W. Fitzgibbon, Javier González Jiménez, Daniel Cremers
- 2017 (modified: 10 Nov 2022)
- CVPR 2017
- Readers: Everyone
Fits Like a Glove: Rapid and Reliable Hand Shape Personalization
David Joseph Tan, Thomas J. Cashman, Jonathan Taylor, Andrew W. Fitzgibbon, Daniel Tarlow, Sameh Khamis, Shahram Izadi, Jamie Shotton
- 2016 (modified: 10 Nov 2022)
- CVPR 2016
- Readers: Everyone
What Shape Are Dolphins? Building 3D Morphable Models from 2D Images.
Thomas J. Cashman, Andrew W. Fitzgibbon
- 2013 (modified: 09 Nov 2022)
- IEEE Trans. Pattern Anal. Mach. Intell.2013
- Readers: Everyone

Co-Authors

View all 40 co-authors