Institute of Communications and Connected Systems


Online Seminar | Towards more efficient video compression and enhancement with machine learning

29 July 2020, 11:30 am–12:30 pm

Images of many video stills hovering around a vector and node representation of a network

Lead R&D Engineer at the BBC, Marta Mrak showcases work being undertaken to exploit Machine Learning for the compression and enhancement of video, exploring the importance of breaking such systems out of the 'black-box'

This event is free.

Event Information

Open to

All | UCL staff | UCL students






Robert Thompson – Institute of Communications and Connected Systems

Towards more efficient video compression and enhancement with machine learning

Machine learning techniques for more efficient video compression and video enhancement have been developed thanks to breakthroughs in deep learning. The new techniques, considered as an advanced form of Artificial Intelligence (AI), bring previously unforeseen capabilities. However, they typically come in the form of resource-hungry black-boxes (overly complex with little transparency regarding the inner workings). Their application can, therefore, be unpredictable and generally unreliable for large-scale use (e.g. in live broadcast).

The aim of the talk is to showcase work that illustrates how understanding the inner workings of Machine Learning (ML) applications can facilitate optimisation of learned models in video processing applications. This means that systems that incorporate them can be used in a more trustworthy manner.  The talk introduces principles for simplification of learned models targeting improved transparency in implementing machine learning for video production and distribution applications. These principles are demonstrated on video compression examples, showing how bitrate savings and reduced complexity can be achieved by simplifying relevant deep learning models.

ICCS Chair

This session will be chaired by: Dr Laura Toni, 

Attending the seminar 

The Seminar will be held on the Zoom platform. Details of how to access Zoom can be found on their website.

Please click this URL to join. https://zoom.us/j/97399330909
Webinar ID: 973 9933 0909
Password: Will be distributed to ICCS members, others are welcome to join and the password can be requested by email iccs@ucl.ac.uk

About the ICCS online seminar series

The ICCS seminar series is designed to bring together members of our community who are currently away from our home in Bloomsbury and distributed across the world. The Seminars are being curated to provide academic exploration and inspiration, offering insights into a range of topics surrounding communications and connected systems. Topics will either explore subjects close to the work of our academics or introduce wider concepts from experts in the global academic and industrial community.

If you wish to suggest a future topic or speaker please use the link below, speakers could be from academia, industry, within ICCS/UCL or from further a field.


The seminar will begin with a presentation aimed at a technical audience, but at a level that will be accessible to those from a range of engineering disciplines. Speakers have been asked to end their presentations with a technical challenge or delving deeper into the content to engage those more invested in the topic. 

The seminar series is designed to offer a compact exploration of ranging topics and is therefore short in format. The presentation will run for 15 - 30 minutes, followed by 15 minutes of questions.
 Suggest a topic or speaker

About the Speaker

Dr Marta Mrak

Lead R&D Engineer at BBC R&D

Dr Marta Mrak (Dipl. Ing.and MSc EE, University of Zagreb, Croatia; PhD, Queen Mary University of London, UK) is a Lead R&D Engineer at BBC R&D and an Honorary Professor at Queen Mary University of London (QMUL), Multimedia and Vision Research Group. Within the IEEE, she is the Chair of Technical Committee on Multimedia Signal Processing, and elected member of IEEE Circuits and Systems Society’s Technical Committee on Multimedia Systems and Applications. Her research is focused on video compression fundamentals, new content experiences and data analytics. Marta serves as General Co-Chair of IEEE ICME 2020 and was the Lead TPC Chair for IEEE ICME 2019. She has organised numerous workshops and special sessions at conferences, in addition to special issues in journals. During 2013-2018, Marta was an Area Editor for Signal Processing Image Communication journal. She was also appointed as Associate Editor for IEEE Trans. on Multimedia in 2018, and an Associate Editor for IEEE Trans. on Image Processing in 2019.

More about Dr Marta Mrak