Research map / Image generation / Autoregressive and latent generation
Adversarially Tuned Scene Generation
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 |
|---|---|---|---|
| Tune scene-generation priors to CityScapes | Synthetic scene data for DeepLab · CityScapes CS_val | global IoU: 49.86 -> 52.14; 2.28 ± 0.34 | Helped |
| Tune scene-generation priors to CamVid | Synthetic scene data for DeepLab · CamVid CV_val | global IoU: 46.42 -> 49.85; +3.42 | Helped |
| Tune scene-generation priors to CityScapes | Synthetic scene data for DeepLab · CamVid CV_val | global IoU: 50.28 (+0.43) | Helped |
| Tune scene-generation priors to CamVid | Synthetic scene data for DeepLab · CityScapes CS_val | global IoU: 39.85 (-6.57) | Hurt |
| Add tuned synthetic data instead of initial synthetic data alongside 10% labeled target samples | DeepLab training · CityScapes CS_val | global IoU: 67.42 -> 70.01; +2.57 | Helped |
| Add tuned synthetic data instead of initial synthetic data alongside 10% labeled target samples | DeepLab training · CamVid CV_val | global IoU: 68.85 -> 70.57; +1.71 | Helped |
| Tune scene-generation priors to CityScapes | Generated scene statistics · CityScapes | KL divergence: 0.57 -> 0.44 | Helped |
Closest papers by meaning
- Generating Triples With Adversarial Networks for Scene Graph Construction (AAAI 2018 · similarity 0.89)
- Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation (CVPR 2020 · similarity 0.88)
- Exploiting Relationship for Complex-scene Image Generation (AAAI 2021 · similarity 0.88)
- Generate Any Scene: Scene Graph Driven Data Synthesis for Visual Generation Training (ICLR 2026 · similarity 0.88)
- Novel View Synthesis with Pixel-Space Diffusion Models (CVPR 2025 · similarity 0.87)
- Unlocking Pre-trained Image Backbones for Semantic Image Synthesis (CVPR 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.