Explorations in the borderlands between vision and language.
In recent years, computational linguistics and computer vision research have developed a sizeable overlap. In this talk I will give a tour of my explorations in this area. These include learning from social media data, grounding weakly supervised text tags with individual image regions, zero-shot learning (How can we use distributional semantic models of words or paragraphs to induce visual recognisers?), learning multi-modal word embeddings and going beyond vector-valued word embeddings to embedding words as probability distributions, deep neural captioning and visual question answering. I aim to present both some mature research and some speculative ideas that I hope will generate fruitful discussion.
Timothy Hospedales is a Reader in Image and Vision Computing at the University of Edinburgh. His research group focuses on lifelong machine learning including transfer and multi-task learning, and cross-modal transfer between language and vision. He has published widely in computer vision and applied machine learning, held grants from EPSRC and EU, and his work received the Best Science Paper award at BMVC 2015. Timothy holds a BA in Computer Science from the University of Cambridge and a PhD in neuroinformatics from the University of Edinburgh.
Add to your calendar
5 May 2017 - Timothy Hospedales: Seminar
Informatics Forum 4.31/4.33