Samuel is an Assistant Professor in the Department of Engineering at the University of Cambridge, a Fellow of the Alan Turing institute and a member of the ELLIS Society.
Andreea-Maria Oncescu is interested in computer vision, natural language applied to audio and video, unsupervised learning. She is based primarily in the Visual Geometry Group at the University of Oxford.
Co-supervised by Samuel Albanie and João F. Henriques.
Gyungin Shin is interested in Computer Vision and Language. He is based primarily in the Visual Geometry Group at the University of Oxford. He is also a Research Scientist at AIFactory Korea
Co-supervised by Samuel Albanie and Weidi Xie.
Jonathan is interested in using computer vision and natural language to interpret satellite imagery. He is part of the Artificial Intelligence for Environmental Risks (AI4ER) CDT.
Co-supervised by Samuel Albanie and Kai Han.
Vishaal Udandarao is interested in multi-modal representation learning (primarily vision-language modelling) in a lifelong learning setting. He is an ELLIS PhD student based primarily at the University of Tuebingen.
Co-supervised by Samuel Albanie and Matthias Bethge.
Tristan is interested in the links between computer vision representations and art historical theories. His research, in digital art history, is funded by a Gates Cambridge scholarship. He is part of the Cambridge Digital Humanities group.
Co-supervised by Samuel Albanie and Leo Impett.
Tingrui He is interested in using language models for code generation to assist researchers and programmers. He is working on a project that involves constructing a new benchmark for evaluating models and improving models' code generation accuracy
Meixi is interested in generation tasks involving vision and language. She is currently working on automatic generation of children story videos. Previously she also conducted research on pure language tasks.
Mingkai Zheng is a Ph.D. student from The University of Sydney. His research interests lie in self-supervised learning, semi-supervised learning, and model pretraining. He is working on few-shot learning with foundation models for augmented scientists.