Antonio Criminisi

Microsoft Research (United Kingdom), Microsoft (United States)

Papers

Total Citations

572

H-Index

About

Antonio Criminisi is a leading figure in computer vision and machine learning, renowned for his pioneering work in interactive image segmentation and 3D scene understanding. His major contributions include the development of the "Geodesic Star Convexity" constraint for interactive image segmentation, a powerful extension of prior shape priors that enables globally optimal solutions for object cut-out—a paper that has garnered over 412 citations. Criminisi is also the driving force behind the groundbreaking "SemanticPaint" system, which revolutionizes 3D scene understanding by allowing users to simultaneously scan and interactively label their environment in real-time, simply by touching objects. This work, with over 160 combined citations, demonstrates his impact in making semantic segmentation accessible and online. His research elegantly bridges the gap between user interaction and automated learning, enabling systems that continuously learn from human input. Criminisi’s achievements have profoundly influenced both academic research and practical applications in augmented reality and robotics, cementing his reputation as an innovator in interactive computer vision.