Multimedia Retrieval Strategies in Videos and the Metaverses Frontier

Upcoming CVC Seminar

Abstract:

Every day, we find ourselves immersed in the age of data. Every hour, vast amounts of media content flood social media and user-generated platforms. For instance, over 500 hours of video are uploaded to YouTube every minute, as of February 2020. Meanwhile, the metaverse is gaining popularity, boasting approximately 400 million monthly users with a continual surge in diverse environments. The proliferation of this extensive and ever-expanding data necessitates advanced and efficient retrieval methods to pinpoint relevant contexts aligned with users' interests.

In this talk, he will discuss the multimedia retrieval task through Natural Language description within two distinct contexts: Text-to-Video Retrieval and Text-to-Metaverse Retrieval. He will present the distinctive features inherent in each context and delve into strategic approaches for developing robust deep learning-based frameworks to help users find content that satisfies their needs. At the end of the presentation, he will explore potentially promising directions in this evolving field.

Short bio:

Giuseppe Serra has been an Associate Professor at the University of Udine, Italy, since November 2019. He received his Ph.D. in Computer Engineering, Multimedia and Telecommunications from the University of Florence, Italy, in 2010. From 2014 to 2016, he was an Assistant Professor at the University of Modena and Reggio Emilia, Italy. He was a visiting researcher at Carnegie Mellon University, Pittsburgh, USA, in 2006, and at Telecom ParisTech/ENST, Paris, in 2010.

His research interests include multimedia, machine learning, and deep learning. He delivered a keynote speech at the International Conference on Machine Vision and Image Processing in 2024. He has been the lead organizer of several international workshops on computer vision and multimedia hosted at leading conferences such as CVPR, ECCV, and ICCV. He has also given tutorials at two international conferences (ICPR '12, CAIP '13). Additionally, he served on the editorial boards of IEEE THMS and ACM TOMM. He has been a Technical Program Committee member for several conferences and workshops, and regularly serves as a reviewer for international conferences and journals such as AAAI, ICML, NIPS, ACL, ECCV, CVPR, IEEE TPAMI, and IEEE TMM. He has published more than 100 publications in the field's most prestigious journals and conferences. He leads the Artificial Intelligence Laboratory at the University of Udine, which consists of more than 10 members, including PhD students and post-docs.