Steven McDonagh
Steven McDonagh is a computer vision and machine learning researcher. He studied AI and computer science at Edinburgh and completed a computer vision doctorate there in 2015. His career includes facial capture at Disney Research, medical image analysis at Imperial College London and research at Huawei’s Noah’s Ark lab. He coauthored Attention U-Net, which adds learned attention gates to medical image segmentation. His later work includes cross-modal depth estimation and naming classes for vision and language models.
- Now at
- University of Edinburgh
- Works
- 83
- Citations
- 7,533
- h-index
- 22
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2008–2015
AI and three-dimensional vision
McDonagh completed an Edinburgh degree in AI and computer science in 2008 and a doctorate in computer vision in 2015. His doctoral work studied density estimation and registration of three-dimensional point sets from multiple views.
-
As of October 2026
Research across imaging settings
He completed postdoctoral work on real-time facial performance capture at Disney Research Zurich and biomedical image analysis at Imperial. He also worked as a senior research scientist at Huawei’s Noah’s Ark lab before Edinburgh.
-
2018
Attention U-Net
McDonagh coauthored Attention U-Net. Its attention gates learn to highlight structures relevant to a segmentation task and suppress other image regions. The paper evaluated the architecture on abdominal CT datasets and released code.
-
2022–2023
Connect visual tasks and language
He coauthored CroMo, a 2022 study of cross-modal learning for single-image depth estimation. In 2023, he coauthored Learning to Name Classes for Vision and Language Models, extending his work into models that combine text and images.
Sources and credits
Photo: ellis.eu; Steven McDonagh profile photograph; creator not stated. License: Not stated by source. Source: ellis.eu. Cropped to a square and resized. Ask us to remove or replace this photo.