Andrew Fitzgibbon
Also known as A W Fitzgibbon, A. Fitzgibbon, A. W Fitzgibbon, A. W. Fitzgibbon, A.W. Fitzgibbon, Andrew W Fitzgibbon, Andrew William Fitzgibbon
Andrew Fitzgibbon is a computer vision researcher whose work connects 3D shape, camera tracking, and human motion recognition. After undergraduate study in Ireland and a PhD at Edinburgh, he leads development of boujou and earns two IEEE Marr Prizes. At Microsoft, he brings synthetic training data into Kinect vision work; a coauthored pose-recognition paper later wins CVPR’s Longuet-Higgins Prize. He is an IAPR Fellow and a Royal Society Fellow, and as of October 2026 he is an Engineering Fellow at Graphcore.
- Now at
- Graphcore
- Works
- 252
- Citations
- 33,684
- h-index
- 63
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
1997
From Ireland to Edinburgh
After undergraduate work at the National University of Ireland, he receives his PhD from Edinburgh University in 1997.
-
1998–2003
Turning geometry into camera paths
His research at Edinburgh and Oxford examines 2D and 3D shape in computer vision. He leads the science and engineering on boujou, software that automatically recovers camera trajectories from 2D images and earns an Emmy for its visual-effects impact. His work earns two IEEE Marr Prizes, including the 2003 ICCV Best Paper for “Image-based Rendering using Image-based Priors.”
-
By 2011
Training computers to read motion
As part of Microsoft’s Kinect team, he helps introduce massive synthetic training data for computer-vision applications. Kinect for Xbox 360 wins the 2011 MacRobert Award.
-
2012
A fellowship for 3D vision
IAPR names him a Fellow for contributions to 3D computer vision research and its applications.
-
2021
A test of time prize
The CVPR Longuet-Higgins Prize recognizes “Real-time human pose recognition in parts from single depth image,” a paper coauthored by Fitzgibbon and seven collaborators.
-
2024
Joining the Royal Society
He becomes a Fellow of the Royal Society in 2024 for contributions that include the boujou 3D camera tracker.
-
As of October 2026
Machine learning meets hardware
At Graphcore, he is an Engineering Fellow. His interests include high-performance programming for hardware and software, with an emphasis on machine-learning models that challenge today’s hardware.
Sources and credits
Photo: graphcore-research.github.io; Andrew Fitzgibbon - Graphcore Research; creator not stated. License: Not stated by source. Source: graphcore-research.github.io. Cropped to a square and resized. Ask us to remove or replace this photo.