Andrea Vedaldi

Andrea Vedaldi is a Professor of Computer Vision and Machine Learning at the University of Oxford's Department of Engineering Science, affiliated with the Visual Geometry Group (VGG). He specializes in unsupervised methods for understanding images and videos, focusing on 3D geometry and semantics. His research bridges foundational AI and practical applications, with contributions to generative models, neural fields, and self-supervised learning.

Education:

• PhD in Computer Science (2008), University of California, Los Angeles
• MSc in Computer Science (2005), UCLA
• BSc in Information Engineering (2003), University of Padua

Research Interests: Unsupervised learning, 3D perception, generative AI, neural rendering, and scalable vision systems. His work emphasizes ethical, responsible AI aligned with ERC-funded projects like UNION (ERC Consolidator Grant).

Key Contributions:

• Co-developer of VLFeat and MatConvNet libraries
• Leader in 3D reconstruction and diffusion models (e.g., CatFree3D)
• Recipient of the PAMI Thomas S. Huang Prize and multiple best paper awards

Grants & Service:

• Principal Investigator on £2.3M ERC Consolidator Grant (UNION)
• Co-organizer of major conferences (ECCV 2020 Program Chair, CVPR 2023 Area Chair)
• Reviewer for top journals/conferences (PAMI, CVPR, NeurIPS)

Labs & Teams: VGG Group at Oxford, collaborating on projects like Meta 3D Gen and Common Objects in 3D (CO3D).