Image Is All for Music Retrieval: Interactive Music Retrieval System Using Images with Mood and Theme Attributes

Abstract

We propose an intuitive image-to-music retrieval (IMR) framework to improve the user experienceon these platforms. The proposed method extracts mood and theme tags by searching for imagesfrom a pre-built database that are similar to a query image and then retrieves music with match-ing tag information. We investigated the system’s effectiveness by comparing participants’satisfac-tion, intention to use, and valence between those who interacted with the system and those whodid not. We also examined whether using mood or theme attributes affected the user-perceivedsuitability of the retrieved music. Results showed that all three variables of the interaction groupwere significantly higher than that of the non-interaction group and that there was no differencein the perceived suitability of music between the mood and theme attributes. Our study con-cludes that image attributes are effective in successful music retrieval and that interaction is a cru-cial factor in designing IMR systems.