Skip to main content
Digital Environment Research Institute (DERI)
  1. Digital Environment Research Institute (DERI)
  2. News & Events
  3. Events 2026
  4. Items

DERI Seminar with Andrew Secker & Juil Sock

When: Thursday, June 8, 2023, 11:00 AM - 12:00 PM
Where: Zoom

Speaker: Dr Andy Secker & Dr Juil Sock, Data Scientists at the BBC

Please join us on Thursday 8th June at 1100hrs when our presenters will be Dr Andy Secker & Dr Juil Sock who are both Data Scientists at the BBC

Together they will give an overview of the BBC's Data science research partnership (of which QMUL is a member), the BBC's uses for data science and the two-way benefits of working in partnership with academic research groups; they will also give us an insight into their on-going project to assist the restoration and colourisation of old images and video at the BBC.

Zoom link: https://qmul-ac-uk.zoom.us/j/81148100921

Dr Andy Secker
Andy is a Lead Data Scientist for BBC R&D, leading the BBC’s Data Science Research Partnership, a research initiative between the BBC and seven UK universities aimed at facilitating and promoting collaborative research in data science and machine learning. Previously Andy led the Language Technology workstream for BBC News Labs focusing on the production of prototypes and tools for the BBC’s World Service news output and leading a portfolio of academic partnerships for non-English language technologies. Andy has a PhD in machine learning for NLP, has co-authored over 30 peer reviewed publications and spent eight years in broadcast engineering for the BBC, contributing to formal standards published by international bodies such as MPEG, DVB and ETSI.

Synopsis
This short introduction will give an overview of the BBC's Data science research partnership (of which QMUL is a member), the BBC's uses for data science and the two-way benefits of working in partnership with academic research groups.

Dr Juil Sock
As the Lead Data Scientist at BBC R&D, Dr Juil Sock holds a Master of Engineering and a PhD in Electrical and Electronic Engineering from Imperial College London, where he conducted research under the guidance of Dr Tae-Kyun Kim. During this time, he developed an efficient learning-based algorithm for detecting and estimating poses of multiple objects. Juil has an extensive publication record, including journal articles, conference papers, and book chapters, and has organized several workshops at conferences. Prior to his role at BBC R&D, Juil worked as a research engineer at a London-based startup.

Synopsis
In this talk, I will present our on-going project to assist the restoration and colourisation of old images and video at the BBC. I will talk about the motivation of the project, how state-of-the-art models are combined and applied, and approaches to mitigate the challenges associated with collecting suitable data. I will conclude by discussing our plans to improve the current work.

Back to top