Distinguished Seminar Series on Data Science & Artificial Intelligence - "Learning Compatible Representation" by Prof. Alberto Del Bimbo
Research Seminar
-
Date
20 Oct 2026
-
Organiser
Department of Computing
-
Time
11:00 - 12:00
-
Venue
PQ305 / Online via Zoom
Speaker
Prof. Alberto Del Bimbo
Summary
Representation learning is a fundamental aspect of deep learning and underpins core tasks such as visual search, retrieval, and recognition. These systems operate by matching images from a gallery set to input query images. They involve using a trained model to encode images from a gallery set into feature representations.
When queries are available, the system retrieves the most similar gallery representations. Early research in these areas has largely considered models that do not change over time. Less attention has been given to scenarios where the availability of new training data requires model updates, or where adopting a more expressive model is needed to enhance performance, which can lead to completely different representations. In these cases, recomputing the feature vectors for all images in the gallery set, a process known as backfilling or re-indexing, becomes essential. However, this can be prohibitively expensive for real-world galleries containing vast amount of data, sometimes even billions of images, or even infeasible if the original data was no more available due to privacy concerns or storage restrictions. As this issue has gained more research attention, learning compatible representations has become a central focus. This approach tackles the difficult problem of learning a new representation model without needing to recalculate gallery features using the updated model.
Our talk will explore the foundational challenges of compatible representation learning and the key role of representation stationarity in this process, along with novel training techniques for developing compatible feature representations through stationarity.
Keynote Speaker
Prof. Alberto Del Bimbo
Emeritus Professor of Computer Engineering
Department of Information Engineering
University of Firenze
Italy
Prof. Alberto Del Bimbo is Emeritus Professor of Computer Engineering at the University of Firenze, Italy, where he served as Director of the Department of Systems and Informatics, Director of the Master’s program in Multimedia, Director of the Media Integration and Communication Center, Deputy Rector for Research and Innovation Transfer and President of the Foundation for Research and Innovation. As a scientist, he authored over 500 scientific publications in Computer Vision, Artificial Intelligence and Multimedia. His current research interests are in compatibility learning, intelligent surveillance and automation and smart environments for cultural heritage. He served as Associate Editor of IEEE TPAMI and IEEE TMM, and was the Editor in Chief of ACM TOMM journal. He was the General Co-Chair of ACMMM (the International Conference on Multimedia) in 2022 and 2010; ICPR (the International Conference on Pattern Recognition) in 2020; ECCV (the European Conference on Computer Vision) in 2012; and IEEE ICMCS (the International Conference on Multimedia Computing and Systems) in 1999. He will be General Co-Chair of ACMMM in London, UK, in 2028. He also served as Program Co-Chair of several other highly ranked scientific conferences. Prof. Del Bimbo has led numerous international and national projects and has been the scientific advisor of institutions and companies worldwide. Presently he is a member of the Scientific Advisory Board of the INSIGHT centre on Artificial Intelligence, Computer Vision and Data Analysis in Ireland. He was the recipient of the ACM SIGMM Award for Outstanding Technical Contributions to Multimedia Computing, Communications and Applications. He is ACM Distinguished Scientist, ELLIS Fellow and IAPR Fellow. From 2021 to 2025, he was the Chair of ACM SIGMM, the ACM Special Interest Group on Multimedia, bringing together the scientific community in Multimedia.
