Prof. Doug Turnbull published the article "You Can Judge an Artist by an Album Cover: Using Images for Music Annotation" in the current issue of the IEEE Multimedia Journal.
"A computer-vision system predicts music genre tags by making use of content-based image analysis, suggesting that we can learn some notion of artists' similarity on the basis of visual appearance alone."
Prof. Turnbull is an assistant professor in the Computer Science Department at Ithaca College. The work was conducted with his former student Janis Libeks. More information about the article can be found at:
http://www.computer.org/portal/web/multimedia
Paper Details:
"You Can Judge an Artist by an Album Cover: Using Images for Music Annotation" by Janis Libeks (U. of Toronto) and Douglas Turnbull (Ithaca College).
IEEE Multimedia, October-December 2011 (vol. 18 no. 4), pp. 30-37
Abstract:
While the perception of music tends to focus on our acoustic listening experience, the image of an artist can play a role in how we categorize (and thus judge) the artistic work. Based on a user study, we show that both album cover artwork and promotional photographs encode valuable information that helps place an artist into a musical context. We also describe a simple computer vision system that can predict music genre tags based on content-based image analysis. This suggests that we can automatically learn some notion of artist similarity based on visual appearance alone. Such visual information may be helpful for improving music discovery in terms of the quality of recommendations, the efficiency of the search process, and the aesthetics of the multimedia experience.
The draft version of the paper can be found at:
http://jimi.ithaca.edu/~dturnbull/Papers/Libeks_ArtistImage_IEEEMM11.pdf
https://www.ithaca.edu/intercom/article.php/20111121080919913