Marta Karczewicz – Advances in video compression (ceremony)


Without my invention what you would be watching
are videos of resolutions maybe 200 by 200 pixels and very few frames per second. The only way to deliver high-quality video
is to compress it. Video compression is based on modelling the
motion between different frames and modelling the correlation between the pixels within
the same frame. Last 20 years I have been working on almost
every part of the modern video codec, contributed hundreds of methods to improve them. Finally, I found my way to USA and I´m currently
working for Qualcomm. This is a company which understands long term
research. When I´m at work I am usually analysing the
problem. But I noticed that the solution really comes
to my mind when I´m outside. When we develop new methods and algorithms
we of course do patent them, but almost immediately we bring them to standardisation forms. We want to inspire other engineers to improve
them further, because standardisation forms foster both collaboration and competition. Currently I´m contributing to VVC standardisation
which is supposed to finish mid of next year and provide another 35% rate reduction over
HEVC. I´m especially proud to work on the standardisation,
because I believe that both, me and my team, contributed to launching the standardisation
effort.

Michael Martin

Leave a Reply

Your email address will not be published. Required fields are marked *

Post comment