Automatic Visual Analysis of Ancient Coins

Sinem Aslan

12 May 2020, 3:00 p.m.

Online Seminar

Go here to participate through Google Meet!

Abstract

Visual analysis of ancient coins has been attracting an increased attention from computer
vision and machine learning community in the last decade. Key ancient numismatics
problems that have been explored by computer vision community so far can be listed as
coin localization in a given cluttered image, coin identification, coin type recognition, issuing
authority recognition, semantic analysis through motif recognition, legend recognition, and
denomination recognition. In this seminar I will introduce key ancient numismatics
problems and I will present our recent work on ancient coin type recognition using Graph
Transduction Games.

Bio sketch

Sinem Aslan is postdoctoral researcher in computer science at DAIS and ECLT of Ca’ Foscari
University of Venice. Her main research activity is in pattern recognition, computer vision
and machine learning, with particular emphasis on image understanding through object-
wise and pixel-wise (semantic segmentation) classification and segmentation where she is
author of 20 publications. Her recent research focuses on computer vision and pattern
recognition within the context of applied work in the areas of cultural heritage. Sinem Aslan
received her Ph.D. from Ege University (Turkey) in 2016. During her Ph.D. studies, she held a
visiting position in BUSIM laboratory at Electrical and Electronics Engineering Department of
Boğaziçi University (Turkey) for two semesters. She was a postdoctoral researcher at IVL of
Milano-Bicocca University (Italy) in 2017.