JOB : Optical Music Recogntion: Graduate Study Opportunity
Vision List Digest:
Article 7,
Volume 18, Issue 24
From: davidb@cs.waikato.ac.nz (David Bainbridge)
Post-Followup: submission@VISLIST.com
I have recently been successful in gaining a grant to pursue a research
area known as Optical Music Recognition (the ability to convert scanned
pages of music into versatile computer-music formats), a topic that
draws heavily on Computer Vision literature. The grant covers a
scholarship for a graduate student to study under my supervision here at
the University of Waikato, New Zealand. Below I have included some
facts and figures about the project -- if you, or someone you know, is
interested in the project please get in touch.
David.
SCHOLARSHIP for PhD or MASTERS DEGREE in COMPUTER MUSIC RESEARCH
A scholarship, worth NZ$14,000 per annum, to support Masters or PhD
study is currently available for a computer music based research
project. The successful candidate will work at the University of
Waikato, New Zealand, in the area of Optical Music Recognition (OMR),
where the aim is to develop software that reliably converts optically
scanned pages of music into versatile machine readable formats. See
below for a more detailed description of the project.
Applicants must have, as minimum, an Honours degree in Computer Science
(or equivalent), and music reading skills, but other than that the
scholarship is unrestricted in who can apply. Because of exchange
agreements, students from France and Germany studying in New Zealand
need pay only local fees (NZ$4000); for other countries, fees are
NZ$17000, however the Department is willing to meet this cost in return
for 6 hours tutoring per week during the teaching year.
Waikato is one of the top Computer Science departments in the country
and has a very successful graduate programme. It is also
internationally one of the leading centres for OMR research. To learn
more about the department, visit its Web site at:
Applications close 21 July.
Prospective candidates should send their current CV, with references, to:
Dr David Bainbridge, Department of Computer Science,
University of Waikato, Hamilton, New Zealand
PROJECT DESCRIPTION
Imagine a computer system that could ``read'' printed music: you could
listen to a piece of written music without any training in musical
notation; a clarinetist could scan a tune and have it transposed
automatically; a soloist could have the computer play an accompaniment
for rehearsal; a music editor could make corrections to an old edition
using a music notation program; or a publisher could reduce an
orchestrated work to a piano part with little fuss, and convert the
piece to Braille with almost no extra work. This is the aim of Optical
Music Recognition (OMR): to convert optically scanned pages of music
into a versatile machine-readable format.
Optical music recognition addresses the problem of musical data
acquisition, the key impediment to the success of the above examples,
however it is a difficult problem that has challenged researchers for
over three decades. Significant achievements have been made over the
years, but no one can yet claim to have developed a robust and versatile
solution. This project aims to surrmount the problem by developing,
from existing expertese in the field, a radically new software framework
that exploits multiple sources of musical knowledge to obtain superiour
levels of accuarcy, whilst keeping the design open so it is extensible
in the set of music notation it can process, thus satisfying the
requirement of versatility.
The project will draw upon the topics of image processing, pattern
recognition, document image analysis, music theory, knowledge
representation schemes, and programming languages. The successful
applicant need not (nor is expected to) be conversant in all these
topics at the start of the project.
http://www.vislist.com