IEM Seminar series: Responsibility by design in audio-visual machine perception


Event details

Date 23.05.2022 15:0016:00  
Speaker Andrea Cavallaro, The Alan Turing Institute, UK
Location Online
Category Conferences - Seminars
Event Language English
Abstract: Machine perception models designed for human facing applications are natural interfaces between the physical and the digital world. These models are transforming the ability of autonomous systems to see, hear and confidently act in a wide range of scenarios. However, the issue of trust is a key challenge for the widespread acceptance of these models, as they capture behaviours that reveal personal information. In this talk I will focus on how we guide machine learning models to learn to serve consensual purposes in human facing applications. I will discuss how we train these models to extract features that are effective for target tasks and prevent unwanted downstream inferences. I will also discuss our recent work that protects selected information by appropriately crafting data perturbations that enable maintaining and even improving sound and image quality.

Bio: Andrea Cavallaro is Turing Fellow at The Alan Turing Institute, the UK National Institute for Data Science and Artificial Intelligence, and Full Professor at Queen Mary University of London, a Russell Group university. He is the founding Director of the Centre for Intelligent Sensing and the Director of Research of the School of Electronic Engineering and Computer Science. He is a Fellow of the International Association for Pattern Recognition for “contributions to image processing and multi-sensor scene understanding,” and a Fellow of the Higher Education Academy. He serves as Editor-in-Chief of Signal Processing: Image Communication, as Senior Area Editor for the IEEE Transactions on Image Processing, as member of the IEEE Video Signal Processing and Communication Technical Committee and as member of the Technical Directions Board of the IEEE Signal Processing Society.

Professor Cavallaro received his Ph.D. in Electrical Engineering from the École polytechnique fédérale de Lausanne (EPFL). He was a Research Fellow with British Telecommunications (BT) and was awarded the Royal Academy of Engineering Teaching Prize; three student paper awards on target tracking and perceptually sensitive coding at IEEE ICASSP in 2005, 2007 and 2009; and the best paper award at IEEE AVSS 2009. He was selected as IEEE Signal Processing Society Distinguished Lecturer (2020-2021) and is the past Chair of the IEEE Image, Video, and Multidimensional Signal Processing Technical Committee (2020-2021). He also served as elected member of the IEEE Multimedia Signal Processing Technical Committee and chair of the Awards committee of the IEEE Signal Processing Society, Image, Video, and Multidimensional Signal Processing Technical Committee.

Practical information

  • General public
  • Free