MusiScene: Leveraging MU-LLaMA for Scene Imagination and Enhanced Video Background Music Generation

TL;DR

MusiScene is a novel model that leverages MU-LLaMA to generate scene-imaginative captions for music, integrating video and music data to improve background music generation with contextually relevant descriptions.

Contribution

The paper introduces MusiScene, a new music captioning model that performs Music Scene Imagination by utilizing a large-scale video-audio dataset and fine-tuning MU-LLaMA, advancing cross-modal scene understanding.

Findings

MusiScene outperforms MU-LLaMA in generating relevant scene captions.

The model enhances video background music generation with contextually appropriate descriptions.

A large-scale dataset of 3,371 video-audio pairs supports the training.

Abstract

Humans can imagine various atmospheres and settings when listening to music, envisioning movie scenes that complement each piece. For example, slow, melancholic music might evoke scenes of heartbreak, while upbeat melodies suggest celebration. This paper explores whether a Music Language Model, e.g. MU-LLaMA, can perform a similar task, called Music Scene Imagination (MSI), which requires cross-modal information from video and music to train. To improve upon existing music captioning models which focusing solely on musical elements, we introduce MusiScene, a music captioning model designed to imagine scenes that complement each music. In this paper, (1) we construct a large-scale video-audio caption dataset with 3,371 pairs, (2) we finetune Music Understanding LLaMA for the MSI task to create MusiScene, and (3) we conduct comprehensive evaluations and prove that our MusiScene is more capable of generating contextually relevant captions compared to MU-LLaMA. We leverage the generated MSI captions to enhance Video Background Music Generation (VBMG) from text.