A study on the active music arrangement using lmage attributes

Abstract

Among different types of media, visual and auditory information are the most important way to perceive the world. People usually judge whether a piece of music or a particular painting is sad or delightful by appreciating it in a concert hall or museum gallery. We also surf the Internet while listening to background music. And many people add their favorite background music as well as updated image files to their blogs in order to represent their daily life and personality. While there has been a lot of research on the relationship between emotion and media, such as music and images, they remain challenging issues because people have different experiences and emotions. We hypothesized that variations of musical and visual image attributes induce emotional changes, and generated a mapping rule between such attributes. We investigated the relationship between emotion and the attributes of such media as music and visual images, and we proposed and evaluated the mapping rule between music and visual image attributes. We also performed preference test with the updated image and generated music, and the average accuracy was 70.4%, which was lower than we expected. However, we were able to find the possibility of a mapping of music and image attributes.