Iro Laina

Iro Laina is a Departmental Lecturer in Computer Vision at the University of Oxford's Visual Geometry Group. She holds a PhD (Dr. rer. nat.) from the Technical University of Munich (TUM), where her dissertation earned the ECVA PhD Award. Her research focuses on unsupervised and language-supervised learning for 3D scene understanding, image/video perception systems, and geometric reconstruction.

Education: PhD in Computer Science (TUM), MSc in Biomedical Computing (TUM), Diploma in Electrical & Computer Engineering (NTUA).

Research Interests:

• 3D Reconstruction and Generation
• Unsupervised Learning
• Multi-View and Video Analysis
• Generative Diffusion Models
• Geometry-Aware Networks

Her recent work emphasizes scalable 3D scene synthesis, training-free methods, and cross-modal fusion with LLMs. Over 15+ publications since 2021 reflect her leadership in geometric deep learning.

Awards: ECVA PhD Award (2020), Recognized in multiple international conferences.

Advising: Mentors DPhil students in creative AI applications (e.g., gameplay design). Active in Oxford's Robotics and Biomedical Engineering networks.

Labs/Tech: Core member of the Visual Geometry Group, collaborating on projects like IMAD2025 with the ZERO Institute.