arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.07063cs.LGcs.AIcs.CVcs.MMcs.SD
收藏

基于视频的音乐生成

Video-based Music Generation

Doctoral Program in Electrical and Computer Engineering(电子与计算机工程博士项目);Faculty of Engineering, University of Porto(葡萄牙波尔图大学工程学院);Faculty of Science and Technology, University of Coimbra(科英布拉大学科学与技术学院);Frost School of Music, University of Miami(迈阿密大学弗罗斯特音乐学院);Faculty of Science and Technology, NOVA University Lisbon(里斯本NOVA大学科学与技术学院);ISEP, Polytechnic of Porto(波尔图理工学院ISEP)

Serkan Sulun

AI总结:

EMSYNC通过视频情感分类、情感驱动音乐生成和时间边界条件,实现了自动化的视频音乐生成,提升了音乐丰富性与情感对齐度。

摘要:

随着互联网上视频内容的快速增长,找到合适的配乐仍是一个重大挑战。本论文提出了EMSYNC(情感和同步),一种快速、免费且自动的解决方案,能够根据输入视频生成定制的音乐,使内容创作者能够在不创作或授权音乐的情况下提升其作品。我们的模型生成的音乐在情感和节奏上与视频同步。EMSYNC的核心组件是一个新的视频情感分类器。通过利用预训练的深度神经网络进行特征提取,并在训练融合层时冻结它们,我们减少了计算复杂性同时提高了准确性。我们通过在Ekman-6和MovieNet上获得最先进的结果来展示方法的泛化能力。另一个关键贡献是一个大规模、带有情感标签的MIDI数据集,用于情感音乐生成。我们随后提出了一个基于情感的MIDI生成器,这是第一个能够根据连续情感值而非离散类别进行条件生成的生成器,使音乐生成能够与复杂的情感内容对齐。为了增强时间同步,我们引入了一种新的时间边界条件方法,称为“边界偏移编码”,将音乐和弦与场景变化对齐。结合视频情感分类、基于情感的音乐生成和时间边界条件,EMSYNC成为了一种完全自动的基于视频的音乐生成器。用户研究显示,它在音乐丰富性、情感对齐、时间同步和整体偏好方面均优于现有方法,为基于视频的音乐生成设定了新的最先进的标准。

英文摘要:

As the volume of video content on the internet grows rapidly, finding a suitable soundtrack remains a significant challenge. This thesis presents EMSYNC (EMotion and SYNChronization), a fast, free, and automatic solution that generates music tailored to the input video, enabling content creators to enhance their productions without composing or licensing music. Our model creates music that is emotionally and rhythmically synchronized with the video. A core component of EMSYNC is a novel video emotion classifier. By leveraging pretrained deep neural networks for feature extraction and keeping them frozen while training only fusion layers, we reduce computational complexity while improving accuracy. We show the generalization abilities of our method by obtaining state-of-the-art results on Ekman-6 and MovieNet. Another key contribution is a large-scale, emotion-labeled MIDI dataset for affective music generation. We then present an emotion-based MIDI generator, the first to condition on continuous emotional values rather than discrete categories, enabling nuanced music generation aligned with complex emotional content. To enhance temporal synchronization, we introduce a novel temporal boundary conditioning method, called "boundary offset encodings," aligning musical chords with scene changes. Combining video emotion classification, emotion-based music generation, and temporal boundary conditioning, EMSYNC emerges as a fully automatic video-based music generator. User studies show that it consistently outperforms existing methods in terms of music richness, emotional alignment, temporal synchronization, and overall preference, setting a new state-of-the-art in video-based music generation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑