Research map / Image generation / Autoregressive and latent generation
Image Generation From Layout
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 image pixel reconstruction loss (ablation) | Layout2Im · Visual Genome | IS: 7.6 $\pm$ 0.2 -> 8.1 $\pm$ 0.1 | Helped |
| add latent code regression loss (ablation) | Layout2Im · Visual Genome | IS: 7.5 $\pm$ 0.1 -> 8.1 $\pm$ 0.1 | Helped |
| add object category classification loss (ablation) | Layout2Im · Visual Genome | Object classification accuracy: 10.06 -> 48.09 | Helped |
| add object category classification loss (ablation) | Layout2Im · Visual Genome | DS: 0.37 $\pm$ 0.11 -> 0.17 $\pm$ 0.09 | Hurt |
| add image adversarial loss (ablation) | Layout2Im · Visual Genome | IS: 7.1 $\pm$ 0.1 -> 8.1 $\pm$ 0.1 | Helped |
| add image adversarial loss (ablation) | Layout2Im · Visual Genome | Object classification accuracy: 56.17 -> 48.09 | Hurt |
| add object adversarial loss (ablation) | Layout2Im · Visual Genome | IS: 7.3 $\pm$ 0.1 -> 8.1 $\pm$ 0.1 | Helped |
| add object adversarial loss (ablation) | Layout2Im · Visual Genome | Object classification accuracy: 57.74 -> 48.09 | Hurt |
| use Layout2Im rather than sg2im (GT Layout) | layout-based image generation · COCO-Stuff | IS: 7.3 $\pm$ 0.1 -> 9.1 $\pm$ 0.1 | Helped |
| use Layout2Im rather than sg2im (GT Layout) | layout-based image generation · Visual Genome | FID: 74.61 -> 31.25 | Helped |
Closest papers by meaning
- Rethinking The Training And Evaluation of Rich-Context Layout-to-Image Generation (NeurIPS 2024 · similarity 0.92)
- OverLayBench: A Benchmark for Layout-to-Image Generation with Dense Overlaps (NeurIPS 2025 · similarity 0.90)
- OverLay++: Dense-Overlap Layout-to-Image Generation Dataset (NeurIPS 2026 · similarity 0.90)
- InstanceAssemble: Layout-Aware Image Generation via Instance Assembling Attention (NeurIPS 2025 · similarity 0.89)
- Build-A-Scene: Interactive 3D Layout Control for Diffusion-Based Image Generation (ICLR 2025 · similarity 0.88)
- Disentangled 3D Scene Generation with Layout Learning (ICML 2024 · similarity 0.88)
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.