Research map / Image generation / Editing and style transfer
Automatic Stage Lighting Control: Is it a Rule-Driven Process or Generative Task?
Zijian Zhao, Dian Jin, Zijing Zhou, Xiaoyu 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 |
|---|---|---|---|
| add skip connection combining music and light embeddings (ablation) | Skip-BART · RPMC-L2 (Value) | RMSE: 68.33 -> 60.74 | Improved |
| add skip connection (ablation) | Skip-BART · RPMC-L2 (Hue) | RMSE: 36.89 -> 36.13 | No clear effect |
| add skip connection (ablation) | Skip-BART · Human eval in-domain | Surprise score: 3.50 -> 3.83 | Improved |
| add skip connection (ablation) | Skip-BART · Human eval in-domain | Smoothness score: 4.43 -> 4.32 | Hurt |
| add pre-training and transfer learning (vs from scratch) (ablation) | Skip-BART · RPMC-L2 (Value) | RMSE: 67.49 -> 60.74 | Improved |
| add light embedding layer (ablation) | Skip-BART · RPMC-L2 (Hue) | RMSE: 51.04 -> 36.13 | Improved |
| add random [MASK] in pre-training (ablation) | Skip-BART · RPMC-L2 (Hue) | RMSE: 49.97 -> 36.13 | Improved |
| add discriminator in pre-training (ablation) | Skip-BART · RPMC-L2 (Hue) | RMSE: 50.40 -> 36.13 | Improved |
| generative Skip-BART instead of rule-based method | Skip-BART · Human eval in-domain | overall score: 2.67 -> 4.35 | Improved |
| compare Skip-BART with real lighting engineers | Skip-BART · Human eval in-domain | overall score: 4.51 vs 4.35 | No clear effect |
Closest papers by meaning
- Stage Light is Sequence$^2$: Multi-Light Control via Imitation Learning (NeurIPS 2026 · similarity 0.93)
- MuseControlLite: Multifunctional Music Generation with Lightweight Conditioners (ICML 2025 · similarity 0.82)
- MusicRL: Aligning Music Generation to Human Preferences (ICML 2024 · similarity 0.81)
- LightIt: Illumination Modeling and Control for Diffusion Models (CVPR 2024 · similarity 0.81)
- MeLFusion: Synthesizing Music from Image and Language Cues using Diffusion Models (CVPR 2024 · similarity 0.81)
- FilmComposer: LLM-Driven Music Production for Silent Film Clips (CVPR 2025 · similarity 0.81)
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.
