Research map / Image generation / Autoregressive and latent generation
Image Generation From Scene Graphs
Read the paperCheck 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 graph convolution to the full model (ablation) | Scene-graph image generation · COCO-Stuff test | Inception score: 4.6 ± 0.1 -> 6.7 ± 0.1 | Helped |
| Add graph convolution to the full model (ablation) | Scene-graph image generation · Visual Genome test | Inception score: 4.2 ± 0.1 -> 5.5 ± 0.1 | Helped |
| Use the scene-graph relationships (ablation) | Scene-graph image generation · COCO-Stuff test | Inception score: 3.7 ± 0.1 -> 6.7 ± 0.1 | Helped |
| Use the scene-graph relationships (ablation) | Scene-graph image generation · Visual Genome test | Inception score: 4.9 ± 0.1 -> 5.5 ± 0.1 | Helped |
| Add both image and object discriminators (ablation) | Scene-graph image generation · COCO-Stuff test | Inception score: 4.8 ± 0.1 -> 6.7 ± 0.1 | Helped |
| Add both image and object discriminators (ablation) | Scene-graph image generation · Visual Genome test | Inception score: 3.6 ± 0.1 -> 5.5 ± 0.1 | Helped |
| Add the object discriminator (ablation) | Scene-graph image generation · COCO-Stuff test | Inception score: 5.6 ± 0.1 -> 6.7 ± 0.1 | Helped |
| Add the object discriminator (ablation) | Scene-graph image generation · Visual Genome test | Inception score: 5.0 ± 0.2 -> 5.5 ± 0.1 | Helped |
| Add the image discriminator (ablation) | Scene-graph image generation · COCO-Stuff test | Inception score: 5.6 ± 0.1 -> 6.7 ± 0.1 | Helped |
| Add the image discriminator (ablation) | Scene-graph image generation · Visual Genome test | Inception score: 5.7 ± 0.3 -> 5.5 ± 0.1 | No clear effect |
| Use ground-truth layouts instead of predicted layouts (ablation) | Scene-graph image generation · COCO-Stuff test | Inception score: 6.7 ± 0.1 -> 7.3 ± 0.1 | Helped |
| Use ground-truth layouts instead of predicted layouts (ablation) | Scene-graph image generation · Visual Genome test | Inception score: 5.5 ± 0.1 -> 6.3 ± 0.2 | Helped |
Closest papers by meaning
- Generating Triples With Adversarial Networks for Scene Graph Construction (AAAI 2018 · similarity 0.93)
- Scene Graph Generation by Iterative Message Passing (CVPR 2017 · similarity 0.93)
- Scene Graph Disentanglement and Composition for Generalizable Complex Image Generation (NeurIPS 2024 · similarity 0.90)
- From Pixels to Graphs: Open-Vocabulary Scene Graph Generation with Vision-Language Models (CVPR 2024 · similarity 0.90)
- Scene Graph-Grounded Image Generation (AAAI 2025 · similarity 0.89)
- DSGG: Dense Relation Transformer for an End-to-end Scene Graph Generation (CVPR 2024 · similarity 0.89)
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.