Speaker: Mr Matthew Aquilina

Date and time: Wednesday 10th November, 12 pm
Venue: Zoom (https://universityofmalta.zoom.us/j/94703348487?pwd=ODI2NGtsR1E0dFcwalE1VmQ2NGZLZz09)

Super-resolution (SR) involves enlarging and enhancing low-resolution images using computational techniques to accurately predict the unseen details of an image (‘Zoom and Enhance’).  The influx of deep learning has pushed SR performance well beyond what was previously thought possible with mathematical/statistical techniques alone.  With today’s convolutional neural networks (CNNs), we can super-resolve artificially degraded (blurred, downsized, etc.) images into incredibly detailed and realistic results, or near-exact replicas of their original high-resolution source.   However, SR is far from being a solved problem, as applying those same networks on real in-the-wild images (taken by a CCTV camera, smartphone, etc.), results in significantly worse quality images, sometimes no better than the standard digital zoom available on our smartphones.  In-the-wild images pass through a large (and typically unknown) number of degrading operations (lens blurring, image compression, noise influx, etc.) before they are eventually saved onto our devices.  Even the largest neural networks are unable to decipher and reverse such a complex web of degradations without guidance.
In the Deep-FIR project, our aim is to introduce new methods for equipping neural networks with the tools to identify and combat various degradations from any input image or video.  Our first step in this direction has been to introduce meta-attention, a lightweight mechanism which allows users to exploit image metadata (attributes) to configure any SR network to identify and counteract various degradations.    

This seminar will delve into the details of meta-attention, as well as highlight and discuss the various research fronts the Deep-FIR project is investigating across both image and video SR.

The Data Science Research Platform (DSRP) at the University of Malta conducts research in the interdisciplinary field of data science. The scope of the group is to use signal processing, machine learning and statistics to develop innovative techniques and to extract useful knowledge from various data sources in an effective manner to benefit the wider public.

For more information about the DSRP, please visit: https://www.um.edu.mt/platform/dsrp

To receive notifications about future events organized by the DSRP, please subscribe to our mailing list: https://groups.google.com/a/um.edu.mt/d/forum/events.dsrp