Optical Music Recognition (OMR) is an important and challenging area within music information retrieval, the accurate detection of music symbols in digital images is a core functionality of any OMR pipeline. In this talk, I will present a novel object detection method, based on synthetic energy maps and the watershed transform, called Deep Watershed Detector (DWD). Our method is specifically tailored to deal with high-resolution images that contain a large number of very small objects and is therefore able to process full pages of written music. We present state-of-the-art detection results of common music symbols in synthetic, scanned and handwritten musical scores.
Ismail Elezi is a Ph.D. student at Ca' Foscari University of Venice under the supervision of professor Marcello Pelillo, in close collaboration with Zurich University of Applied Sciences, working with professor Thilo Stadelmann. He received a Bachelor degree in Computer Science from the University of Prishtina in Kosovo, and a master degree in Computer Science from Ca' Foscari University of Venice where he worked on Graph Theory. His Ph.D. studies are in the field of Deep Learning with the emphasis on the use of contextual information in Deep Learning. During 2017 and 2018, he spent a year in Zurich University of Applied Sciences where he co-developed a novel object recognition method. His research interests are Machine Learning, Deep Learning, and Computer Vision while he finds fascinating Deep Reinforcement Learning. He is preparing a MOOC in Deep Learning for DataCamp.