Page 17 - IO_magazine

In conversation with

COMPUTER SCIENCE

CLOSE TO REAL WORLD

Pablo Cesar, group leader of the Distributive and Interactive Systems group at Centrum
Wiskunde & Informatica (CWI) and Associate Professor at TU Delft, is the recipient of the
2020 Netherlands Prize for ICT Research. His work is about modelling and controlling complex
collections of media objects. ‘We need human-centred computing, as otherwise, we may get

the wrong type of intelligent machines’.

By Bennie Mols
Image Elodie Burrillon

How would you characterise your work? You study systems in a realistic setting.
‘The two main pillars of my group’s work are What is the added value?
affective computing and human-centred multi - ‘My aim is to bring our research as close as
media systems. First, you need to understand possible to the real world. Lab results are
and model the context, intention and state of the essential, but labs never show the actual high
users. After that, you can optimise the multimedia noise in data that you experience in the real
system for that intention and context. Examples world. For example, we are planning to put
of the latter include algorithms for better com- an installation in the Netherlands Institute for
pressing multimedia signals or for orchestrating Sound and Vision to gather data from users
different media streams in an optimal manner.’ accessing the archives. Via these data, we
can understand the real experience of visitors
Can you give some concrete examples? instead of the experience in some artificial
‘We are exploring social virtual reality as a new laboratory context. In the past, we have
medium for remote communication and collabo- followed this methodology in-the-wild,
ration. Social virtual reality consists of highly for example during a theatre play at the
immersive and multi-sensory VR-systems. For National Theatre of China.’
example, we work together with the Netherlands
Institute for Sound and Vision to investigate how What drives you in your work?
to enrich remote access to their collection, and we ‘We need human-centred computing, as
are trying to start a project with a Dutch hospital otherwise, we may get the wrong type of intelli-
to study how to optimise remote consultation, gent machines. The Netherlands needs to better
based on SocialVR principles.’ support human-computer interaction research
in computer science, as already happens in the
What are the big challenges in your field? USA or Germany. I hope that the Netherlands
‘One big challenge is understanding of the activity Prize for ICT Research helps to further improve
and context of users in real life. This problem can the recognition of this important field. In the
only be studied by combining various scientific future, multimedia systems will be intelligent
disciplines. The second big challenge is to opti- and data-centric. It is up to us to make them
mise the quality of the experience by transmitting human-centred and empathic as well.’
and orchestrating the important bits. The ques-
tion is not about how fast the bits are transported,
but how well they are utilised. The end goal must
always be to provide the best possible experience
for the user.’

17