Shahram Izadi
Microsoft Research (United Kingdom), Microsoft (United States)
Papers
8
Total Citations
1,402
H-Index
7
About
Shahram Izadi is a pioneering researcher at the intersection of computer vision, mixed reality, and 3D scene understanding, whose work has fundamentally advanced how machines perceive and reconstruct the physical world. He is perhaps best known for BundleFusion, a landmark system enabling real-time, globally consistent 3D reconstruction of large-scale scenes — a breakthrough that addressed the longstanding challenge of drift in pose estimation without requiring hours of offline processing. Published in 2017, BundleFusion alone has accumulated over 1,000 citations across its various versions, underscoring its profound influence on both robotics and mixed reality communities. Izadi's research extends into semantic scene understanding, with contributions like SemanticPaint, an innovative interactive system allowing users to simultaneously scan and semantically label 3D environments through natural touch-based interaction, accumulating over 160 citations. His work on incremental dense semantic stereo fusion further pushed the boundaries of large-scale scene reconstruction. Beyond perception, Izadi has explored novel interaction paradigms, including "mechanical hijacking" for simplified ubiquitous computing deployment and lightweight VR tracking through ThirdLight. Collectively, his research has shaped modern approaches to spatial computing, making him a central figure in advancing immersive and intelligent environments.
Research Focus
Key Achievements
Top Papers
- 1BundleFusion680 citations · 2017
- 2BundleFusion245 citations · 2017
- 3
- 4
- 5SemanticPaint83 citations · 2015
- 6SemanticPaint: Interactive 3D Labeling and Learning at your Fingertips77 citations · 2015
- 7Mechanical hijacking14 citations · 2011
- 8ThirdLight2 citations · 2017