On music genre classification via compressive sampling

Bob L. Sturm

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

15 Citationer (Scopus)
975 Downloads (Pure)


Recent work \cite{Chang2010} combines low-level acoustic features
and random projection (referred to as ``compressed sensing'' in \cite{Chang2010})
to create a music genre classification system showing
an accuracy among the highest reported
for a benchmark dataset.
This not only contradicts previous findings
that suggest low-level features are inadequate for
addressing high-level musical problems,
but also that a random projection of features
can improve classification.
We reproduce this work and resolve these contradictions.
TidsskriftInternational Conference on Multimedia and Expo
StatusUdgivet - 2013
Begivenhed2012 IEEE Conference on Multimedia & Expo - San Jose, USA
Varighed: 15 jul. 201319 jul. 2013


Konference2012 IEEE Conference on Multimedia & Expo
BySan Jose


Dyk ned i forskningsemnerne om 'On music genre classification via compressive sampling'. Sammen danner de et unikt fingeraftryk.