This is a classic multi-label classification problem. Prediction of movie genres is an intriguing problem that has several applications in designing recommendation systems for the audiences, analyzing movie box office performance and understanding the theme of the movie to list some. This work adds to our understanding of music’s use in multi-modal contexts and offers the potential for future inquiry into human affective experiences. Furthermore, we provide software code to replicate this study at.
We investigate the interaction between musical and visual features with a cross-modal analysis, and do not find compelling evidence that music characteristic of a certain genre implies low-level visual features associated with that genre. We examine the best-performing MIR feature model through permutation feature importance (PFI), determining that mel-frequency cepstral coefficient (MFCC) and tonal features are most indicative of musical differences between genres. We use these models to compare handcrafted music information retrieval (MIR) features against VGGish audio embedding features, finding similar performance with the top-performing architectures.
We construct supervised neural network models with various pooling mechanisms to predict a film’s genre from its soundtrack. This study investigates such phenomena through a quantitative evaluation of music that is associated with different film genres. For instance, composers may evoke passion in a romantic scene with lush string passages or inspire fear throughout horror films with inharmonious drones. The results of the study support the notion that high intensity movies (i.e., Action and Horror) have musical cues that are measurably different from the musical scores for movies with more measured expressions of emotion (i.e., Drama and Romance).įilm music varies tremendously across genre in order to bring about different responses in an audience. Both pair-wise genre classification and classification with all four genres was performed using support vector machines (SVM) in a ten-fold cross-validation test. For this study, a database of film scores from 98 movies was collected containing instrumental (non-vocal) music from 25 romance, 25 drama, 23 horror, and 25 action movies. The intent of this study was to provide a preliminary understanding on a new database for the impact of timbral and select rhythm features in characterizing the differences among movie genres based on their film scores. While it is clear that the full emotional effect of a movie scene is carried through the successful interpretation of audio and visual information, music still carries a significant impact for interpretation of the director's intent and style.
0 kommentar(er)
