Research map / Video / Video-language understanding
VMChill: A Dataset for Fine-Grained Visual-Musical Synergy
Xiaowei Chi, Zeyue Tian, Jialiang Chen, Wei Xue
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 |
|---|---|---|---|
| fine-tune VideoCrafter-2.0 on VMChill-2M | VideoCrafter-2.0 · VBench | subject consistency: 96.85 -> 98.62 | Improved |
| fine-tune VideoCrafter-2.0 on VMChill-2M | VideoCrafter-2.0 · VBench | motion smoothness: 97.73 -> 98.33 | Improved |
| fine-tune VideoCrafter-2.0 on VMChill-2M | VideoCrafter-2.0 · VBench | overall consistency: 28.23 -> 25.33 | Hurt |
| fine-tune VideoCrafter-2.0 on VMChill-2M | VideoCrafter-2.0 · VBench | temporal style: 25.84 -> 24.61 | Hurt |
| fine-tune VideoCrafter-2.0 on VMChill-2M | VideoCrafter-2.0 · VBench | sum of 9 dimensions: 64.45 -> 64.57 | No clear effect |
| fine-tune VideoCrafter-2.0 on VMChill-2M | VideoCrafter-2.0 · VBench | image quality: 67.22 -> 69.78 | Improved |
| train Video-LLaMA on VMChill-2M instead of official fine-tuned weights | Video-LLaMA · VMChill-Test | CIDEr: 2.45 -> 24.79 | Improved |
| train Video-LLaMA on VMChill-2M instead of official fine-tuned weights | Video-LLaMA · VMChill-Test | BLEU-4: 3.94 -> 5.59 | Improved |
| train Video-LLaMA on VMChill-2M instead of official fine-tuned weights | Video-LLaMA · VMChill-Test | METEOR: 14.05 -> 13.83 | Hurt |
| fine-tune Stable-Audio-Open on VMChill-2M | Stable-Audio-Open · MusicCaps | FAD: 3.23 -> 2.40 | Improved |
| fine-tune Stable-Audio-Open on VMChill-2M | Stable-Audio-Open · MusicCaps | KL: 1.51 -> 1.20 | Improved |
| use DiT trained from scratch on VMChill-2M (V2M-DiT) instead of VidMuse | video-to-music · V2M-Bench | FAD: 2.46 -> 2.51 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- VidMuse — VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling (CVPR 2025) · better on 0 of 2 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
- HarmonySet: A Comprehensive Dataset for Understanding Video-Music Semantic Alignment and Temporal Synchronization (CVPR 2025 · similarity 0.88)
- VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling (CVPR 2025 · similarity 0.87)
- V2Meow: Meowing to the Visual Beat via Video-to-Music Generation (AAAI 2024 · similarity 0.87)
- Customized Condition Controllable Generation for Video Soundtrack (CVPR 2025 · similarity 0.86)
- LLark: A Multimodal Instruction-Following Language Model for Music (ICML 2024 · similarity 0.86)
- Mirasol3B: A Multimodal Autoregressive Model for Time-Aligned and Contextual Modalities (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.
