Research map / 3D vision / 3D generation
Event-Driven Storytelling with Multiple Lifelike Humans in a 3D Scene
Donggeun Lim, Jinseok Bae, Inwoo Hwang, Seungmin Lee, Hwanhee Lee, Young Min Kim
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 event-driven planning (narrator and event parser) (ablation) | GPT-4o planner · Multi-agent planning benchmark | success rate: 0.82 -> 0.9 | Improved |
| add event-driven planning (ablation) | GPT-4o mini planner · Multi-agent planning benchmark | success rate: 0.6 -> 0.74 | Improved |
| add event-driven planning (ablation) | Llama-3.1-8B planner · Multi-agent planning benchmark | success rate: 0.31 -> 0.35 | Improved |
| add event-driven planning, scene-state tag (ablation) | GPT-4o planner · Multi-agent planning benchmark (SS) | success rate: 0.77 -> 0.92 | Improved |
| add event-driven planning (ablation) | DeepSeek-V3 planner · Multi-agent planning benchmark | success rate: 0.82 -> 0.83 | No clear effect |
| use context-centric scene description instead of object list (ablation) | GPT-4o planner · Multi-agent planning benchmark | success rate: 0.51 -> 0.9 | Improved |
| use context-centric scene description instead of scene graph (ablation) | GPT-4o mini planner · Multi-agent planning benchmark | success rate: 0.49 -> 0.74 | Improved |
| use context-centric scene description instead of scene graph (ablation) | Llama-3.1-70B planner · Multi-agent planning benchmark | success rate: 0.72 -> 0.72 | No clear effect |
| use area-conditioned position sampling instead of direct coordinates (ablation) | GPT-4o planner · PI benchmark | success rate: 0.46 -> 0.94 | Improved |
| use area-conditioned position sampling instead of direct coordinates (ablation) | Qwen2.5-72B planner · PI benchmark | success rate: 0.14 -> 1.0 | Improved |
| use area-conditioned position sampling instead of direct coordinates (ablation) | Llama-3.1-8B planner · PI benchmark | success rate: 0.14 -> 0.86 | Improved |
| replace text scene description with vision-generated description (ablation) | GPT-4o planner · Multi-agent planning benchmark | success rate: 0.9 -> 0.68 | Hurt |
Closest papers by meaning
- Talk2Event: Grounded Understanding of Dynamic Scenes from Event Cameras (NeurIPS 2025 · similarity 0.87)
- Generating Human Motion in 3D Scenes from Text Descriptions (CVPR 2024 · similarity 0.86)
- Motion-Agent: A Conversational Framework for Human Motion Generation with LLMs (ICLR 2025 · similarity 0.86)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.86)
- ContextualStory: Consistent Visual Storytelling with Spatially-Enhanced and Storyline Context (AAAI 2025 · similarity 0.86)
- The Language of Motion: Unifying Verbal and Non-verbal Language of 3D Human Motion (CVPR 2025 · 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.
