Research map / Image generation / Autoregressive and latent generation
Probabilistic Neural Programmed Networks for Scene Generation
Zhiwei Deng, Jiacheng Chen, YIFANG FU, Greg Mori
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 |
|---|---|---|---|
| use PNP-Net instead of DC-GAN/PixelCNN/PoE-VAE baselines | PNP-Net · CLEVR-G | OBJ-A detector mAP: 0.176 -> 0.737 (vs DC-GAN) | Improved |
| use PNP-Net instead of baselines | PNP-Net · CLEVR-G zero-shot | OBJ-A detector mAP: 0.332 -> 0.734 (vs DC-GAN) | Improved |
| use PNP-Net instead of baselines | PNP-Net · ColorMNIST | OBJ-A detector mAP: 0.318 -> 0.363 (vs PixelCNN) | Improved |
| use PNP-Net instead of PixelCNN/DC-GAN | PNP-Net · ColorMNIST | OBJ-N detector mAP: 0.990 -> 0.981 (vs DC-GAN) | Hurt |
| use gated PoE instead of plain PoE in combine operator (ablation) | PNP-Net · ColorMNIST | detector score: 2-3% boost | Improved |
| train on up-to-4-object scenes instead of up-to-8 and test on up-to-8 (ablation) | PNP-Net · CLEVR-G 128x128 up to 8 objects | OBJ-A detector mAP: 0.567 -> 0.518 | Hurt |
Closest papers by meaning
- Compositional Scene Understanding through Inverse Generative Modeling (ICML 2025 · similarity 0.89)
- Generative Scene Graph Networks (ICLR 2021 · similarity 0.88)
- Deep Generative Probabilistic Graph Neural Networks for Scene Graph Generation (AAAI 2020 · similarity 0.88)
- Scene Graph Disentanglement and Composition for Generalizable Complex Image Generation (NeurIPS 2024 · similarity 0.88)
- GENESIS: Generative Scene Inference and Sampling with Object-Centric Latent Representations (ICLR 2020 · similarity 0.88)
- Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis (CVPR 2025 · similarity 0.87)
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.
