Research map / 3D vision / 3D generation
External Knowledge Enhanced 3D Scene Generation from Sketch
Zijie Wu, Mingtao Feng, Yaonan Wang, He Xie, Weisheng Dong, Bo Miao, Ajmal Mian
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 knowledge-enhanced graph reasoning (knowledge base) to the sketch condition (ablation) | SEK · 3D-FRONT ablation | FID: 33.29 -> 23.97 | Improved |
| add sketch condition to knowledge-only model (ablation) | SEK · 3D-FRONT ablation | FID: 32.26 -> 23.97 | Improved |
| add spectrum-filter block (ablation) | SEK denoiser · 3D-FRONT ablation | FID: 25.83 -> 23.97 | Improved |
| use ViT instead of ResNet50 sketch encoder when knowledge is present (ablation) | SEK · 3D-FRONT ablation | FID: 24.68 -> 23.97 | No clear effect |
| use ViT instead of ResNet50 sketch encoder without knowledge (ablation) | SEK · 3D-FRONT ablation | FID: 34.76 -> 33.29 | Improved |
| use a knowledge base built on 3D-FRONT instead of empty for ScanNet (ablation) | SEK · ScanNet knowledge transfer | FID: 44.27 -> 33.81 | Improved |
| retrain knowledge base on ScanNet instead of transferring from 3D-FRONT (ablation) | SEK · ScanNet knowledge transfer | FID: 33.81 -> 33.47 | No clear effect |
| add knowledge-enhanced sketch condition to ATISS | ATISS · 3D-FRONT Bedroom sketch-based generation | FID: 21.33 -> 18.47 | Improved |
| add knowledge-enhanced sketch condition to Sceneformer | Sceneformer · 3D-FRONT Bedroom sketch-based generation | FID: 37.21 -> 24.68 | Improved |
| use SEK instead of DiffuScene for scene completion | scene completion · 3D-FRONT Bedroom scene completion | FID: 27.32 -> 21.84 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DiffuScene — DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (CVPR 2024) · better on 7 of 7 shared results
- ATISS — ATISS: Autoregressive Transformers for Indoor Scene Synthesis (NeurIPS 2021) · 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
- DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (CVPR 2024 · similarity 0.89)
- Denoising Diffusion via Image-Based Rendering (ICLR 2024 · similarity 0.89)
- LT3SD: Latent Trees for 3D Scene Diffusion (CVPR 2025 · similarity 0.88)
- R3CD: Scene Graph to Image Generation with Relation-Aware Compositional Contrastive Control Diffusion (AAAI 2024 · similarity 0.88)
- Coherent 3D Scene Diffusion From a Single RGB Image (NeurIPS 2024 · similarity 0.88)
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025 · 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.
