Research map / 3D vision / 3D generation
FunFact: Building Probabilistic Functional 3D Scene Graphs via Factor-Graph Reasoning
Zhengyu Fu, René Zurbrügg, Kaixian Qu, Marc Pollefeys, Marco Hutter, Hermann Blum, Zuria Bauer
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 factor graph reasoning (cardinality and proximity factors) (ablation) | FunFact · FunThor | edge F1 (%): 31.1 -> 38.7 | Improved |
| add factor graph reasoning (ablation) | FunFact · FunThor | ECE (all edges): 0.70 -> 0.36 | Improved |
| add factor graph reasoning (ablation) | FunFact · FunThor ambiguous relations | ECE (ambiguous): 0.45 -> 0.07 | Improved |
| add factor graph reasoning (ablation) | FunFact · FunThor | edge recall (%): 53.4 -> 49.3 | Hurt |
| add hierarchical object and part proposal (ablation) | FunFact · FunThor | edge F1 (%): 19.8 -> 38.7 | Improved |
| add hierarchical object and part proposal (ablation) | FunFact · FunThor | interactive element Recall@3: 41.8 -> 69.5 | Improved |
| add hierarchical object and part proposal (ablation) | FunFact · FunThor | object Recall@3: 68.9 -> 68.2 | No clear effect |
| FunFact instead of OpenFunGraph on FunGraph3D triplets | functional scene graph · FunGraph3D | overall triplet Recall@5: 29.8 -> 48.7 | Improved |
| FunFact instead of OpenFunGraph on SceneFun3D | functional scene graph · SceneFun3D | overall triplet Recall@5: 60.4 -> 41.0 | Hurt |
| FunFact instead of OpenFunGraph on FunThor triplets | functional scene graph · FunThor | overall triplet Recall@3: 15.1 -> 54.1 | Improved |
Closest papers by meaning
- Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spaces (CVPR 2025 · similarity 0.91)
- SceneFun3D: Fine-Grained Functionality and Affordance Understanding in 3D Scenes (CVPR 2024 · similarity 0.89)
- SceneScaffold: Active Scene-State Construction for Unified 3D Scene Understanding (NeurIPS 2026 · similarity 0.89)
- TB-HSU: Hierarchical 3D Scene Understanding with Contextual Affordances (AAAI 2025 · similarity 0.88)
- Functionality Understanding and Segmentation in 3D Scenes (CVPR 2025 · similarity 0.88)
- Open3DSG: Open-Vocabulary 3D Scene Graphs from Point Clouds with Queryable Objects and Open-Set Relationships (CVPR 2024 · 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.
