Research map / Video / Video generation
FoleyDirector: Fine-Grained Temporal Steering for Video-to-Audio Generation via Structured Scripts
You Li, Dewei Zhou, Fan Ma, Fu Li, Dongliang He, Yi Yang
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 |
|---|---|---|---|
| add Structured Temporal Scripts via Temporal Script Attention (ablation) | MMAudio · DirectorBench (overall) | F1: 0.1311 -> 0.4252 | Improved |
| add interleaved RoPE to STS fusion (ablation) | FoleyDirector · DirectorBench (overall) | F1: 0.4252 -> 0.4389 | Improved |
| add Bi-Frame Sound Synthesis (ablation) | FoleyDirector · DirectorBench (overall) | F1: 0.4389 -> 0.4819 | Improved |
| add Bi-Frame on the complex reannotated subset (ablation) | FoleyDirector · DirectorBench (reannotated subset) | F1: 0.4178 -> 0.4613 | Improved |
| shorten STS segments from 1s to 0.5s at inference (ablation) | FoleyDirector · DirectorBench (overall) | F1: 0.4819 -> 0.5197 | Improved |
| lengthen STS segments from 1s to 2s at inference (ablation) | FoleyDirector · DirectorBench (overall) | F1: 0.4819 -> 0.4646 | Hurt |
| add STS control to MMAudio | MMAudio · VGGSound-Director | FD_VGG: 1.45 -> 1.17 | Improved |
| add STS control to MMAudio | MMAudio · VGGSound-Director | FD_PANN / FD_PaSST: 8.23 -> 8.60; 92.80 -> 94.27 | Mixed |
| disable SG-TFM at inference (Ours w/o STS) (ablation) | FoleyDirector · VGGSound-Director | FD_PANN / DeSync: FD_PANN 8.58 vs MMAudio 8.23; DeSync 0.438 vs 0.439 | No clear effect |
Closest papers by meaning
- Tri-Ergon: Fine-Grained Video-to-Audio Generation with Multi-Modal Conditions and LUFS Control (AAAI 2025 · similarity 0.89)
- Foley-Flow: Coordinated Video-to-Audio Generation with Masked Audio-Visual Alignment and Dynamic Conditional Flows (CVPR 2025 · similarity 0.88)
- Audio-Sync Video Generation with Multi-Stream Temporal Control (NeurIPS 2025 · similarity 0.88)
- Read, Watch and Scream! Sound Generation from Text and Video (AAAI 2025 · similarity 0.87)
- ThinkSound: Chain-of-Thought Reasoning in Multimodal LLMs for Audio Generation and Editing (NeurIPS 2025 · similarity 0.86)
- Tell What You Hear From What You See - Video to Audio Generation Through Text (NeurIPS 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.
