Research map / 3D vision / 3D generation
Forest2Seq: Revitalizing Order Prior for Sequential Indoor Scene Synthesis
Qi Sun, Hang Zhou, Wengang Zhou, Li Li, Houqiang Li
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 tree-guided BFS ordering instead of single random ordering (ablation) | Forest2Seq · 3D-FRONT living room | FID: 49.4 -> 36.1 | Improved |
| use scene forest instead of single scene tree (ablation) | Forest2Seq · 3D-FRONT living room | KL: 7.90 -> 5.90 | Improved |
| use BFS instead of DFS traversal of the forest (ablation) | Forest2Seq · 3D-FRONT living room | FID: 40.5 -> 35.2 | Improved |
| use multiple random orderings instead of a single random ordering (ablation) | autoregressive scene model · 3D-FRONT living room | FID: 49.4 -> 43.3 | Improved |
| add absolute positional encoding (ablation) | Forest2Seq · 3D-FRONT living room | KL: 11.1 -> 5.9 | Improved |
| absolute versus relative positional encoding (ablation) | Forest2Seq · 3D-FRONT living room | KL: 6.2 vs 5.9 | No clear effect |
| add denoising with 0.05 mask/noise rate (ablation) | Forest2Seq · 3D-FRONT living room | KL: 6.4 -> 5.9 | Improved |
| raise mask/noise rate to 0.2 (ablation) | Forest2Seq · 3D-FRONT living room | KL: 5.9 -> 12.2 | Hurt |
| use MEDC (add GIoU term) instead of plain Euclidean clustering (ablation) | scene tree parsing · 3D-FRONT scene parsing | AHD: 0.53 -> 0.84 | 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 2 of 2 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
- InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior (ICLR 2024 · similarity 0.90)
- Hierarchically-Structured Open-Vocabulary Indoor Scene Synthesis with Pre-trained Large Language Model (AAAI 2025 · similarity 0.89)
- SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent (NeurIPS 2025 · similarity 0.88)
- S-INF: Towards Realistic Indoor Scene Synthesis via Scene Implicit Neural Field (AAAI 2025 · similarity 0.87)
- DeBaRA: Denoising-Based 3D Room Arrangement Generation (NeurIPS 2024 · similarity 0.87)
- CASAGPT: Cuboid Arrangement and Scene Assembly for Interior Design (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.
