Silvia Zottin
PhD student, University of Udine
Silvia Zottin is a Ph.D. student at the Artificial Vision and Machine Learning (AVML) lab at University of Udine. She received her B.Sc. in Computer Science from Ca' Foscari University in 2018 and completed her M.Sc. degree Summa Cum Laude in Multimedia Communication and Information Technologies at University of Udine in 2021.
Her research focuses on computer vision, anomaly detection, image segmentation, and machine learning in limited data contexts. She has published in national and international journals and conferences, contributing to the advancement of her field.
She is actively involved in the research community, serving as one of the organizers of the Competition on Few-Shot and Many-Shot Layout Segmentation of Ancient Manuscripts (SAM) at ICDAR 2024, as well as co-organizing the VisionDocs workshop at both WACV 2025 and ICCV 2025.
Workshop
Document Analysis and Understanding in the Era of Deep Learning
with Axel De Nardin
Written communication has always been a key aspect of human civilization since ancient history. The ability to understand written information through vision in an automated way sparked the interest of computer scientists for a long time, making it one of the oldest fields in computer vision. Stemming from this interest a wide variety of tasks have been identified in this field going from document analysis to text understanding in the wild.
This lecture aims at providing an overview of the recent advancements in the field of Document Analysis and Understanding, introducing the opportunities opened up by the employment of recent Deep Learning technologies as well as the challenges that still characterize this research area despites our best efforts.
The lecture will be organized in two sections, a theoretical one which aims at providing a better understanding of the key aspects of document analysis and understanding. And a hands-on practical session which will provide an opportunity to implement a simple document analysis framework.
