5 papers
Diff-V2M: A Hierarchical Conditional Diffusion Model with Explicit Rhythmic Modeling for Video-to-Music Generation
Shulei Ji, Zihao Wang, Jiaxing Yu +4
Video-to-music (V2M) generation aims to create music that aligns with visual content. However, two main challenges persist in existing methods: (1) the lack of explicit rhythm mode…
Assessing Data Replication in Symbolic Music via Adapted Structural Similarity Index Measure
Shulei Ji, Zihao Wang, Le Ma +2
AI-generated music may inadvertently replicate samples from the training data, raising concerns of plagiarism. Similarity measures can quantify such replication, thereby offering s…
Losing is for Cherishing: Data Valuation Based on Machine Unlearning and Shapley Value
Le Ma, Shirao Yang, Zihao Wang +4
The proliferation of large models has intensified the need for efficient data valuation methods to quantify the contribution of individual data providers. Traditional approaches, s…
A Comprehensive Survey on Generative AI for Video-to-Music Generation
Shulei Ji, Songruoyao Wu, Zihao Wang +2
The burgeoning growth of video-to-music generation can be attributed to the ascendancy of multimodal generative models. However, there is a lack of literature that comprehensively…
MetaBGM: Dynamic Soundtrack Transformation For Continuous Multi-Scene Experiences With Ambient Awareness And Personalization
Haoxuan Liu, Zihao Wang, Haorong Hong +5
This paper introduces MetaBGM, a groundbreaking framework for generating background music that adapts to dynamic scenes and real-time user interactions. We define multi-scene as va…