Research map / Video / Video generation
Lay2Story: Extending Diffusion Transformers for Layout-Togglable Story Generation
Ao Ma, Jiasong Feng, Ke Cao, Jing Wang, Yun Wang, Quanwei Zhang, Zhanjie 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 subject branch (ablation) | Lay2Story · Lay2Story-Bench | FID: 110.58 -> 26.71 | Improved |
| add subject branch (ablation) | Lay2Story · Lay2Story-Bench | Recall@1: 0.1733 -> 0.7012 | Improved |
| add reference image concatenation (ablation) | Lay2Story · Lay2Story-Bench | FID: 50.27 -> 26.71 | Improved |
| add reference image concatenation (ablation) | Lay2Story · Lay2Story-Bench | Human preference: 0.4781 -> 0.7561 | Improved |
| add masked 3D self-attention (ablation) | Lay2Story · Lay2Story-Bench | FID: 66.14 -> 26.71 | Improved |
| add masked 3D self-attention (ablation) | Lay2Story · Lay2Story-Bench | Recall@1: 0.3274 -> 0.7012 | Improved |
| inject layout conditions (subject positions and captions) at inference | Lay2Story · Lay2Story-Bench | FID: 35.82 -> 26.71 | Improved |
| inject layout conditions at inference | Lay2Story · Lay2Story-Bench | DreamSim: 0.1602 -> 0.1324 | Improved |
| inject layout conditions at inference | Lay2Story · Lay2Story-Bench | Recall@1: 0.6376 -> 0.7012 | Improved |
| inject layout conditions at inference | Lay2Story · Lay2Story-Bench | inference time (s): 13.63 -> 14.02 | Hurt |
| replace FLUX.1-dev IP-Adapter with Lay2Story w/o layout | Lay2Story w/o lc · Lay2Story-Bench | inference time (s): 61.38 -> 13.63 | Improved |
| replace FLUX.1-dev IP-Adapter with Lay2Story w/o layout | Lay2Story w/o lc · Lay2Story-Bench | FID: 33.18 -> 35.82 | Hurt |
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.
- StoryDiffusion — StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation (NeurIPS 2024) · better on 2 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
- Storybooth: Training-Free Multi-Subject Consistency for Improved Visual Storytelling (ICLR 2025 · similarity 0.88)
- Directed Diffusion: Direct Control of Object Placement through Attention Guidance (AAAI 2024 · similarity 0.88)
- Storynizor: Consistent Story Generation via Inter-Frame Synchronized and Shuffled ID Injection (AAAI 2025 · similarity 0.87)
- One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt (ICLR 2025 · similarity 0.87)
- ContextualStory: Consistent Visual Storytelling with Spatially-Enhanced and Storyline Context (AAAI 2025 · similarity 0.87)
- SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency (NeurIPS 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.
