Research map / Video / Video generation
Diff-V2M: A Hierarchical Conditional Diffusion Model with Explicit Rhythmic Modeling for Video-to-Music Generation
Shulei Ji, Zihao Wang, Jiaxing Yu, Xiangyuan Yang, Shuyu Li, Songruoyao Wu, Kejun Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use low-resolution ODF instead of low-res mel-spectrogram as rhythm representation (ablation) | Diff-V2M · Mixed test set | FAD: 2.0376 -> 2.0175 | Improved |
| use low-resolution ODF instead of tempogram (ablation) | Diff-V2M · Mixed test set | FAD: 2.3163 -> 2.0175 | Improved |
| add post-attention FiLM with feature selection over weighted fusion (ablation) | Diff-V2M · Mixed test set | FAD: 2.3625 -> 1.5175 | Improved |
| apply FiLM after rather than before attention (ablation) | Diff-V2M · Mixed test set | FAD: 2.0812 -> 2.1340 | No clear effect |
| add rhythmic features (ablation) | Diff-V2M · Mixed test set | FAD: 1.8264 -> 1.5175 | Improved |
| add rhythmic features (ablation) | Diff-V2M · Mixed test set | ImageBind score: 0.1893 -> 0.1812 | Hurt |
| add emotional features (ablation) | Diff-V2M · Mixed test set | FAD: 1.6761 -> 1.5175 | Improved |
| add visual rhythm inputs (scene onset, visual beats) to rhythm predictor (ablation) | Diff-V2M · Mixed test set | FAD: 2.2177 -> 1.5175 | Improved |
| jointly train rhythm predictor and generator (ablation) | Diff-V2M · Mixed test set | FAD: 1.8787 -> 1.5175 | Improved |
| jointly train rhythm predictor and generator (ablation) | Diff-V2M · V2M-Bench | FAD: 3.3245 -> 1.7612 | Improved |
| add scheduled conditioning (ablation) | Diff-V2M · Mixed test set | FAD: 1.6159 -> 1.5175 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- VidMuse — VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling (CVPR 2025) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- V2Meow: Meowing to the Visual Beat via Video-to-Music Generation (AAAI 2024 · similarity 0.90)
- Customized Condition Controllable Generation for Video Soundtrack (CVPR 2025 · similarity 0.88)
- VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling (CVPR 2025 · similarity 0.88)
- MoMu-Diffusion: On Learning Long-Term Motion-Music Synchronization and Correspondence (NeurIPS 2024 · similarity 0.88)
- HarmonySet: A Comprehensive Dataset for Understanding Video-Music Semantic Alignment and Temporal Synchronization (CVPR 2025 · similarity 0.87)
- MeLFusion: Synthesizing Music from Image and Language Cues using Diffusion Models (CVPR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
